所有标签 - AI Papers Daily

所有标签

LLM ×552 GRPO ×328 RAG ×267 Benchmark ×261 LLM Agents ×220 Reinforcement Learning ×186 Agent ×149 LoRA ×137 Semantic ID ×134 RLVR ×122 Multi-Agent ×119 RL ×111 LLM Agent ×111 Evaluation ×110 Multimodal ×106 MoE ×104 Generative Recommendation ×104 LLM agents ×90 benchmark ×89 Tool Use ×84 Knowledge Distillation ×78 Reasoning ×78 On-Policy Distillation ×73 Sequential Recommendation ×70 evaluation ×68 SFT ×67 VLM ×66 MLLM ×64 RLHF ×63 Agent Memory ×55 KV cache ×55 Generative Retrieval ×52 Self-Distillation ×51 Memory ×49 Personalization ×49 Retrieval ×48 Contrastive Learning ×47 Credit Assignment ×47 reasoning ×45 Chain-of-Thought ×44 reinforcement learning ×43 DPO ×42 LLM Evaluation ×42 Agentic RL ×41 Dense Retrieval ×41 LLM Reasoning ×40 Efficiency ×39 Knowledge Graph ×39 KV Cache ×38 World Model ×38 Interpretability ×38 Post-training ×37 Continual Learning ×37 LLM evaluation ×37 Long Context ×36 Video Generation ×36 Speculative Decoding ×36 Long-Horizon ×35 E-commerce ×35 LLM Inference ×34 Quantization ×34 Synthetic Data ×34 Diffusion Models ×33 Training-Free ×33 Fine-tuning ×33 Multimodal LLM ×33 LLM Training ×33 multimodal ×32 Self-Improvement ×32 Agent Evaluation ×30 Code Generation ×30 Calibration ×30 GUI Agent ×29 Agent Harness ×29 multi-agent ×29 Self-Evolution ×28 CTR Prediction ×28 MultiAgent ×28 on-policy distillation ×28 training-free ×27 MCP ×27 E-commerce Search ×27 Flow Matching ×26 Distillation ×26 Reranking ×26 Multi-hop QA ×26 Transformer ×26 Latent Reasoning ×25 Multi-Agent Systems ×25 Token Efficiency ×25 calibration ×25 test-time scaling ×24 Agentic Search ×24 In-Context Learning ×24 LLM-as-a-Judge ×24 Policy Optimization ×24 PPO ×24 quantization ×24 LLM inference ×24 Sparse Attention ×23 Model Merging ×23 Scaling Laws ×23 SWE-bench ×23 Recommendation ×23 Curriculum Learning ×23 interpretability ×23 post-training ×23 Preference Optimization ×23 Survey ×22 Test-Time Scaling ×22 PEFT ×22 RQ-VAE ×22 Reward Model ×22 Training-free ×22 Recommender Systems ×22 agent ×22 Search Agent ×22 Recursive Self-Improvement ×22 LLM training ×21 Inference Efficiency ×21 dense retrieval ×21 long-context ×21 LLM4Rec ×21 Training Efficiency ×21 knowledge distillation ×21 self-distillation ×20 LLM reasoning ×20 Mixture-of-Experts ×20 distillation ×20 Multimodal Agent ×20 Agentic AI ×20 scaling laws ×20 Catastrophic Forgetting ×20 Vision-Language Models ×20 Agent Skills ×20 tool use ×20 Vision-Language Model ×20 Mathematical Reasoning ×19 synthetic data ×19 Process Reward ×19 Collaborative Filtering ×19 Safety ×18 Deep Research ×18 Inference Acceleration ×18 GraphRAG ×18 Hallucination ×18 Generalization ×18 Agent Training ×18 Context Management ×18 mechanistic interpretability ×18 On-policy Distillation ×18 personalization ×18 Long-horizon ×17 LLM-as-Judge ×17 sparse attention ×17 Reward Hacking ×17 Long-term Memory ×17 LLM serving ×17 credit assignment ×17 Data Synthesis ×16 Agentic RAG ×16 Exploration ×16 Video Understanding ×16 speculative decoding ×16 retrieval ×16 Image Generation ×16 LLM agent ×16 LLM Safety ×16 BM25 ×16 Verification ×16 Domain Adaptation ×15 Beam Search ×15 Multimodal Retrieval ×15 Multilingual ×15 Post-Training ×15 Prompt Engineering ×15 Muon ×15 Graph Neural Networks ×15 Skill Library ×15 Preference Alignment ×15 LLM-as-judge ×15 reward hacking ×15 coding agents ×15 chain-of-thought ×15 ANN ×15 Routing ×14 Training Stability ×14 Linear Attention ×14 multilingual ×14 Embodied AI ×14 Self-Evolving ×14 Agent Benchmark ×14 Robustness ×14 Human-AI Collaboration ×14 alignment ×14 Scientific Discovery ×14 LLM Judge ×14 test-time compute ×14 Diffusion Language Models ×14 Coding Agent ×14 Reliability ×14 Parallel Decoding ×14 Cold-Start ×13 Multimodal RAG ×13 User Simulation ×13 Temporal Reasoning ×13 long-horizon ×13 fine-tuning ×13 Coding Agents ×13 Pretraining ×13 Late Interaction ×13 agent memory ×13 Uncertainty Estimation ×13 LLM post-training ×13 Prompt Optimization ×13 model compression ×13 A-B Testing ×13 abstention ×13 LLM Post-training ×13 world model ×12 Long-Context ×12 Gated DeltaNet ×12 efficiency ×12 KL Divergence ×12 Explainability ×12 Causal Inference ×12 Faithfulness ×12 long context ×12 World Models ×12 recommender systems ×12 Mechanistic Interpretability ×12 Multimodal Recommendation ×12 cold-start ×12 Reward Modeling ×12 self-improvement ×12 Program Synthesis ×12 Retrieval-Augmented Generation ×12 Residual Quantization ×12 Computer-Use Agents ×12 Zero-shot ×12 Diversity ×12 generalization ×12 Efficient Inference ×12 RoPE ×12 Evidence Grounding ×12 Harness ×12 Query Expansion ×12 policy optimization ×11 Test-time Scaling ×11 VLA ×11 Multimodal Agents ×11 Semantic IDs ×11 Foundation Model ×11 Process Supervision ×11 coding agent ×11 Spatial Reasoning ×11 Multimodal Reasoning ×11 Long-Horizon Agents ×11 LLM-as-a-judge ×11 contrastive learning ×11 AdamW ×11 Self-Evolving Agents ×11 Alignment ×11 Prompt Injection ×11 Inverted Index ×11 ColBERT ×11 Hybrid Retrieval ×11 Self-Consistency ×11 activation steering ×11 Skill Evolution ×11 PTQ ×11 Online Learning ×11 Tree Search ×11 reranking ×11 Privileged Information ×11 Agentic Reasoning ×10 Diffusion Transformer ×10 Model Compression ×10 KV cache compression ×10 vLLM ×10 Planning ×10 Visual Reasoning ×10 Attribution ×10 Tool Calling ×10 Vector Quantization ×10 LLM Fine-tuning ×10 Conversational AI ×10 Provenance ×10 training efficiency ×10 prompt optimization ×10 Grounding ×10 Multi-Task Learning ×10 Constrained Decoding ×10 Industrial Recommendation ×10 vector search ×10 curriculum learning ×10 efficient inference ×10 Failure Analysis ×10 LLM judge ×10 continual learning ×10 instruction following ×10 reliability ×10 Foundation Models ×10 FlashAttention ×9 Compression ×9 diffusion models ×9 Matryoshka ×9 OCR ×9 Math Reasoning ×9 Conformal Prediction ×9 Uncertainty ×9 On-Policy ×9 Agentic Workflow ×9 Ranking ×9 Vector Search ×9 Security ×9 Load Balancing ×9 prompt injection ×9 CLIP ×9 multi-agent systems ×9 Taxonomy ×9 Uncertainty Quantification ×9 Mixture of Experts ×9 Attention ×9 Discrete Diffusion ×9 agent evaluation ×9 ReAct ×9 Image Editing ×9 hallucination detection ×9 Multi-Token Prediction ×9 Evolutionary Search ×9 Hypergraph ×9 LLM Pretraining ×9 Test-Time Training ×9 Tool Retrieval ×9 latent reasoning ×9 LLMs ×9 On-policy distillation ×9 LLM Distillation ×9 LLM Compression ×9 Product Quantization ×9 robustness ×9 safety ×9 Optimization ×9 pretraining ×9 Multi-task Learning ×9 Query Rewriting ×9 Industrial Deployment ×9 Text-to-SQL ×9 multi-hop QA ×9 Multi-Teacher Distillation ×9 recursive self-improvement ×9 Harness Evolution ×9 Web Agent ×8 Trajectory Analysis ×8 LVLM ×8 Diffusion Distillation ×8 Multi-agent ×8 Web Agents ×8 Training ×8 Environment Synthesis ×8 CoT ×8 in-context learning ×8 Retrieval-Augmented ×8 SLM ×8 flow matching ×8 uncertainty quantification ×8 Multi-Agent System ×8 LLM pretraining ×8 Information Retrieval ×8 Multi-modal ×8 LLM Alignment ×8 confidence calibration ×8 Self-Play ×8 Text Generation ×8 Conversational Recommendation ×8 ASR ×8 Rubric ×8 Memory Management ×8 inference efficiency ×8 Self-Supervised Learning ×8 inference optimization ×8 Reward Design ×8 Scaling ×8 LLM Reranking ×8 Long-tail ×8 Computer Use ×8 Hyperparameter Optimization ×8 Feature Interaction ×8 instruction tuning ×8 Harness Optimization ×8 Reward Shaping ×8 LLM Serving ×8 agentic search ×8 Scaling Law ×8 sequential recommendation ×8 Cost Efficiency ×8 on-policy ×8 Looped Transformer ×8 reasoning models ×8 Hybrid Attention ×8 Test-time Compute ×8 Language Modeling ×8 LLM fine-tuning ×8 On-Policy Self-Distillation ×8 Contextual Bandit ×8 long-term memory ×8 3D Generation ×7 Pre-training ×7 Policy Gradient ×7 self-evolving ×7 POMDP ×7 code generation ×7 DAG ×7 verification ×7 Reproducibility ×7 AI Agents ×7 GEO ×7 Visual Grounding ×7 Time Series Forecasting ×7 Open-Source ×7 Failure Diagnosis ×7 video generation ×7 Cross-Attention ×7 Abstention ×7 Off-Policy Evaluation ×7 Pruning ×7 Function Calling ×7 Cold Start ×7 Optimizer ×7 Memory Evolution ×7 catastrophic forgetting ×7 conformal prediction ×7 Popularity Bias ×7 Skill Optimization ×7 Multi-Agent RL ×7 dataset ×7 ranking ×7 memory ×7 Bias ×7 Variance Reduction ×7 Importance Sampling ×7 uncertainty estimation ×7 Token Compression ×7 Cross-Encoder ×7 Adversarial Training ×7 Agent Safety ×7 model merging ×7 Test-Time Compute ×7 Parametric Memory ×7 Long-Term Memory ×7 Computer-Use Agent ×7 vision-language model ×7 Verifiable Rewards ×7 KV-cache ×7 training data ×7 Open Source ×7 Vision-Language ×7 Long-Context Reasoning ×7 Hard Negative Mining ×7 Thompson Sampling ×7 Long-context ×7 context compression ×7 Formal Verification ×7 HNSW ×7 transformer ×7 eviction ×7 Hierarchical Memory ×7 Learning to Rank ×7 Simulation ×7 Feature Engineering ×7 Rejection Sampling ×7 Teacher-Student ×7 selective prediction ×7 Hallucination Mitigation ×7 Software Engineering ×7 knowledge graph ×7 linear attention ×7 Inference Optimization ×7 Adversarial Robustness ×7 open-source ×7 cost efficiency ×7 User Simulator ×7 foundation model ×7 User Modeling ×7 model routing ×7 attention ×7 routing ×7 Sequence Modeling ×7 Controllable Generation ×7 LLM safety ×7 probing ×7 LLM Routing ×7 Robot Manipulation ×7 importance sampling ×7 Multi-hop Reasoning ×7 Meta-Agent ×7 confidence estimation ×7 FP8 ×7 Attention Heads ×6 Time Series ×6 conversational AI ×6 Dynamic Routing ×6 image generation ×6 RecSys ×6 AutoML ×6 scalability ×6 Human-in-the-loop ×6 Sycophancy ×6 Plug-and-Play ×6 positional encoding ×6 Trust Region ×6 Scalability ×6 Human-in-the-Loop ×6 Active Learning ×6 Evidence Graph ×6 Efficient Training ×6 Episodic Memory ×6 Autonomous Research ×6 Privacy ×6 NVFP4 ×6 Counterfactual ×6 Rubric Reward ×6 Multi-Objective RL ×6 pruning ×6 Offline RL ×6 Query Generation ×6 compression ×6 Masked Diffusion ×6 bias ×6 Looped Transformers ×6 LLM alignment ×6 MaxSim ×6 InfoNCE ×6 Multimodal Embedding ×6 Reflection ×6 OOD Generalization ×6 Multi-turn Interaction ×6 debiasing ×6 Hallucination Reduction ×6 Instruction Following ×6 Language Agents ×6 Synthetic Environments ×6 Persona ×6 Sparse Retrieval ×6 Sparse Autoencoder ×6 Context Selection ×6 Graph Retrieval ×6 Risk Control ×6 SPLADE ×6 Latent Memory ×6 MDP ×6 Multi-turn ×6 Hypernetwork ×6 human-in-the-loop ×6 MCMC ×6 attention heads ×6 steering ×6 Regularization ×6 Distributed Training ×6 QLoRA ×6 long-context inference ×6 GUI Agents ×6 Safety Alignment ×6 Representation Alignment ×6 KV Cache Compression ×6 Cold-start ×6 LLM RL ×6 Streaming Video ×6 Learned Sparse Retrieval ×6 ANNS ×6 Root Cause Analysis ×6 diffusion language models ×6 autoregressive ×6 generative models ×6 Optimal Transport ×6 Agentic Memory ×6 Tokenization ×6 context management ×6 hidden states ×6 Memory Efficiency ×6 optimization ×6 Skill Distillation ×6 tool calling ×6 Terminal-Bench ×6 KL Regularization ×6 LLM Post-Training ×6 QAT ×6 popularity bias ×6 pass@k ×6 counterfactual ×6 agent training ×6 test-time adaptation ×6 Adaptive Routing ×6 GNN ×6 LLM Security ×6 token efficiency ×6 Chunking ×6 Agent RL ×6 diffusion model ×5 Compositional Reasoning ×5 Procedural Memory ×5 PRM ×5 unsupervised learning ×5 Unified Architecture ×5 Transformer Architecture ×5 Token Pruning ×5 Search ×5 Embedding ×5 image editing ×5 Procedural Knowledge ×5 Knowledge Base ×5 Memory Augmentation ×5 Context Engineering ×5 Bradley-Terry ×5 Unified Model ×5 domain adaptation ×5 Knowledge Graphs ×5 Multimodal Alignment ×5 Inference ×5 explainability ×5 GPU Kernel ×5 Triton ×5 Agentic System ×5 Hierarchical Retrieval ×5 label noise ×5 Knowledge Injection ×5 Multi-Agent Simulation ×5 optimizer ×5 entropy ×5 Multimodal Memory ×5 Gating ×5 Data Quality ×5 Debiasing ×5 multi-turn ×5 inference acceleration ×5 Online RL ×5 Inference-time ×5 AI Agent ×5 DiT ×5 Agentic LLM ×5 visual reasoning ×5 BPE ×5 Latent Communication ×5 entropy collapse ×5 Semantic Retrieval ×5 Sponsored Search ×5 Dual Encoder ×5 Long-Horizon Agent ×5 generative recommendation ×5 Mixed-Precision ×5 Tool Selection ×5 data synthesis ×5 Adaptive Inference ×5 SSM ×5 agent harness ×5 Denoising ×5 privacy ×5 Proactive Agents ×5 RL fine-tuning ×5 multimodal agent ×5 Advantage Estimation ×5 e-commerce search ×5 LLM-based Recommendation ×5 on-policy self-distillation ×5 Efficient Attention ×5 Agentic Systems ×5 Draft Model ×5 Early Stopping ×5 Fairness ×5 Music Recommendation ×5 Data Augmentation ×5 Multimodal Fusion ×5 Long-Horizon Reasoning ×5 Mobile Agent ×5 Confidence Calibration ×5 online learning ×5 LLM routing ×5 Activation Steering ×5 Agent Security ×5 Reranker ×5 Theory of Mind ×5 sparse autoencoders ×5 data augmentation ×5 GUI Grounding ×5 Cross-Domain ×5 bi-encoder ×5 web agents ×5 video understanding ×5 Hallucination Detection ×5 Multimodal Learning ×5 Self-Supervised ×5 Semantic Alignment ×5 Conversational Search ×5 AI agents ×5 Long Video Understanding ×5 Agentic Retrieval ×5 user modeling ×5 Discrete Representation ×5 Execution Feedback ×5 faithfulness ×5 SVD ×5 Multi-turn Agent ×5 Search Agents ×5 Adaptive Retrieval ×5 discrete diffusion ×5 preference optimization ×5 Self-Reflection ×5 retrieval-augmented ×5 Agentic Framework ×5 Architecture ×5 Query Reformulation ×5 agentic RL ×5 Skill Routing ×5 Overfitting ×5 Distribution Shift ×5 Policy Learning ×5 Ablation Study ×5 on-device ×5 Deep Research Agent ×5 variance reduction ×5 Industrial RecSys ×5 Causal Reasoning ×5 long-context LLM ×5 Autonomous Agents ×5 Context Compaction ×5 multimodal retrieval ×5 Text-to-Image ×5 Model Evaluation ×5 Test-Time Adaptation ×5 LLM Reranker ×5 self-improving agents ×5 Streaming ×5 Memory Optimization ×5 policy gradient ×5 Guardrails ×5 Benchmarking ×5 inference ×5 Multi-Objective ×5 JEPA ×5 Skill Retrieval ×5 Frozen LLM ×5 CTR prediction ×5 OSWorld ×5 long-horizon agents ×5 Omni-modal ×4 Action Prediction ×4 State Tracking ×4 language model ×4 Mid-Training ×4 text-to-image ×4 Listwise Reranking ×4 Document Retrieval ×4 Collaboration ×4 Failure Attribution ×4 Error Correction ×4 3D Gaussian Splatting ×4 Training Dynamics ×4 MMD ×4 RL Post-training ×4 Temporal Consistency ×4 Policy Distillation ×4 Evidence Selection ×4 Information Gain ×4 Text-to-Image Generation ×4 vision-language models ×4 visual grounding ×4 Auditing ×4 Dataset ×4 Closed-Loop ×4 planning ×4 orchestration ×4 EHR ×4 Best-of-N ×4 mutual information ×4 Serving ×4 Architecture Search ×4 GPTQ ×4 Context-Aware ×4 Meta-Learning ×4 Hierarchical Reasoning ×4 Parallel Reasoning ×4 Confidence Estimation ×4 Multi-turn Dialogue ×4 Self-supervised ×4 Newton-Schulz ×4 Flat Minima ×4 Reasoning Distillation ×4 vision-language ×4 variational inference ×4 Monitoring ×4 Federated Learning ×4 Privacy-Preserving ×4 Self-Correction ×4 Video Editing ×4 Working Memory ×4 Tool-Use ×4 Pareto Front ×4 computer-use agents ×4 human-AI interaction ×4 VLM Judge ×4 Unsupervised Learning ×4 DAPO ×4 Zero-Shot ×4 UCB ×4 low-rank ×4 denoising ×4 human-AI collaboration ×4 embedding compression ×4 End-to-End ×4 Scheduling ×4 Latency Optimization ×4 Data Selection ×4 Influence Function ×4 Lean ×4 Progressive Disclosure ×4 On-policy ×4 skill evolution ×4 adversarial training ×4 Decision Support ×4 Human-AI Interaction ×4 SAE ×4 AgenticRL ×4 MLLMs ×4 chunking ×4 Structured Pruning ×4 training stability ×4 streaming ×4 hybrid retrieval ×4 memory consolidation ×4 Position Bias ×4 Search Efficiency ×4 trust region ×4 HyDE ×4 benchmarking ×4 Ensemble ×4 Online A-B Test ×4 distribution shift ×4 Pareto Frontier ×4 Co-Training ×4 External Memory ×4 Visual Memory ×4 reranker ×4 DCI ×4 data selection ×4 Reverse KL ×4 recommendation ×4 compositional generalization ×4 training dynamics ×4 transformers ×4 Self-correction ×4 Heterogeneous Data ×4 Positional Encoding ×4 Skill Learning ×4 Data Curation ×4 inference-time scaling ×4 Model Distillation ×4 cross-encoder ×4 Interactive Evaluation ×4 POI Recommendation ×4 prompt compression ×4 Pareto Optimization ×4 Industrial Recommender ×4 data curation ×4 rubric-based RL ×4 Intent Inference ×4 Spatial Intelligence ×4 Relevance Modeling ×4 Generative Models ×4 meta-learning ×4 Memorization ×4 sample efficiency ×4 multimodal LLM ×4 representation learning ×4 DPP ×4 Visual Token Pruning ×4 E-Commerce Search ×4 User Profiling ×4 offline RL ×4 Entropy ×4 Query Recommendation ×4 multi-modal ×4 IVF ×4 attribution ×4 GBDT ×4 Early Exit ×4 search agent ×4 Instruction Tuning ×4 Information Theory ×4 Persistent Memory ×4 optimal transport ×4 Agent Architecture ×4 pre-training ×4 self-supervised learning ×4 Delayed Feedback ×4 Consistency ×4 Reinforcement Fine-Tuning ×4 Long-Horizon Tasks ×4 LLM Planning ×4 VQA ×4 hybrid attention ×4 NoPE ×4 Pretraining Efficiency ×4 Long-Document QA ×4 Adaptive Computation ×4 OPSD ×4 Self-Verification ×4 Diffusion Language Model ×4 Decision Making ×4 Two-Tower ×4 Multilingual Reasoning ×4 Bias Mitigation ×4 user study ×4 FP4 ×4 human alignment ×4 fairness ×4 TTFT ×4 Model Routing ×4 BEIR ×4 safety alignment ×4 clustering ×4 Lean 4 ×4 LLM deployment ×4 self-consistency ×4 Workflow ×4 Task Planning ×4 Intervention ×4 position bias ×4 LLM compression ×4 Question Answering ×4 In-context Learning ×4 Cost-Aware ×4 debate ×4 Adapter ×4 Red Teaming ×4 Self-evolving ×4 Data Mixture ×4 taxonomy ×4 Data Agents ×4 memory management ×4 Representation Learning ×4 Dynamic Programming ×4 Query Understanding ×4 Skill Selection ×4 MCTS ×4 Retrieval Evaluation ×4 spectral analysis ×4 Zeroth-Order Optimization ×4 Vision Transformer ×4 e-commerce ×4 late interaction ×4 diffusion ×4 parallel decoding ×4 Evaluation Protocol ×4 linear probe ×4 Transformers ×4 Visual Analytics ×4 Majority Voting ×4 reasoning distillation ×4 Industrial Search ×4 verifier ×4 Metacognition ×4 grounding ×4 Pseudo-Relevance Feedback ×4 Memory-Augmented ×4 CVR Prediction ×4 brand recommendation ×4 Entity Resolution ×4 A-B Test ×4 Doubly Robust ×4 KBQA ×4 model evaluation ×4 SWE-Bench ×4 residual stream ×4 Data Generation ×4 transfer learning ×4 Probing ×4 Operations Research ×4 red-teaming ×4 Sample Efficiency ×4 spatial reasoning ×4 Co-Evolution ×4 Cross-encoder ×4 compute-optimal ×4 self-play ×4 Weak-to-Strong ×4 Sparse Routing ×4 LLM Orchestration ×4 Recommendation Systems ×4 OPD ×4 Adversarial Attack ×4 Terminal Agent ×4 training ×4 RRF ×4 Causal Intervention ×4 Experience Reuse ×4 Compute Efficiency ×4 Long-horizon Agent ×4 Multilingual LLM ×4 LLM distillation ×4 Terminal Agents ×4 tabular foundation models ×4 LLM Memory ×4 Context Compression ×4 CKA ×4 ARC-AGI ×4 prefix caching ×4 Expert Routing ×4 Query Performance Prediction ×4 Embodied Agent ×4 Enterprise AI ×4 tool-using agents ×4 Iterative Retrieval ×4 Process Evaluation ×3 reward model ×3 CLI ×3 imitation learning ×3 hallucination ×3 Evidence-Grounded ×3 Residual Connections ×3 Bayesian optimization ×3 AutoResearch ×3 Self-Refinement ×3 VAE ×3 Few-shot Learning ×3 Forecasting ×3 μP ×3 Bayesian Inference ×3 sampling ×3 Robotics ×3 Lifelong Learning ×3 Multi-agent systems ×3 Trajectory ×3 Skill Reuse ×3 Sample Complexity ×3 Diffusion Transformers ×3 test-time compute scaling ×3 Few-Shot ×3 Token-Level Gating ×3 exploration ×3 Eval ×3 TTS ×3 Token Reduction ×3 active learning ×3 regression ×3 epistemic uncertainty ×3 real-time ×3 Parameter-Efficient ×3 Token Weighting ×3 Adaptive Reasoning ×3 XGBoost ×3 Evidence Aggregation ×3 Frozen Model ×3 Machine Unlearning ×3 adversarial attack ×3 Tokenizer ×3 bootstrap ×3 Inference-time Scaling ×3 Verifiable Reward ×3 Asynchronous ×3 Codebook ×3 BERT ×3 Transfer Learning ×3 Literature Retrieval ×3 Graph Learning ×3 Hierarchical Decomposition ×3 Agent Self-Improvement ×3 security ×3 Multi-Agent LLM ×3 Industrial Recommender System ×3 RL Training ×3 Self-improvement ×3 omni-modal ×3 audio-visual ×3 Evolution ×3 Hindsight Distillation ×3 Structured Data ×3 Environment Scaling ×3 Multi-Objective Optimization ×3 Dynamic Weighting ×3 workflow ×3 Hindsight ×3 GenRec ×3 evolutionary search ×3 Token Credit Assignment ×3 latency reduction ×3 Experience Distillation ×3 Learning Rate Schedule ×3 function calling ×3 Diffusion ×3 Vision Foundation Model ×3 adapter ×3 Delta Rule ×3 Bias Correction ×3 Long-term memory ×3 Data Engine ×3 Reasoning Efficiency ×3 Structured Reasoning ×3 Entity Linking ×3 Skill Extraction ×3 Negative Transfer ×3 Meta-Skill ×3 Sandbox ×3 Sim-to-Real ×3 agent skills ×3 vocabulary ×3 Multi-hop Retrieval ×3 self-correction ×3 Conversational Agent ×3 Structured Generation ×3 Candidate Generation ×3 prompt sensitivity ×3 tool-calling ×3 diversity ×3 Open-Ended Tasks ×3 Query Decomposition ×3 Listwise Ranking ×3 terminal agents ×3 latency ×3 multi-token prediction ×3 Long-horizon Agents ×3 Multi-Vector Retrieval ×3 RMSNorm ×3 trajectory evaluation ×3 skill routing ×3 Video Retrieval ×3 Knowledge Boundary ×3 Plackett-Luce ×3 agent safety ×3 Generative Reranking ×3 Minecraft ×3 Harness Engineering ×3 multi-objective optimization ×3 off-policy ×3 NLI ×3 Proactive Agent ×3 Self-Organization ×3 Domain Specialization ×3 Quality-Diversity ×3 Tool-Augmented ×3 gradient alignment ×3 failure attribution ×3 Variational Inference ×3 Rubrics ×3 Self-Improving Agents ×3 A-B test ×3 approximate nearest neighbor ×3 Query Refinement ×3 Graph Memory ×3 Autoresearch ×3 SVG ×3 Human-Agent Interaction ×3 Stability-Plasticity ×3 Forgetting ×3 contextual bandits ×3 Pinterest ×3 Parameter-Efficient Fine-Tuning ×3 model scaling ×3 Ensembling ×3 Training Recipe ×3 AUROC ×3 Value Function ×3 reward modeling ×3 Reasoning Traces ×3 Sparse Reward ×3 self-verification ×3 collaborative filtering ×3 Contextual Bandits ×3 Decoding ×3 sparse retrieval ×3 cross-domain recommendation ×3 implicit feedback ×3 Inference-Time Scaling ×3 Trustworthy AI ×3 computer-use ×3 GUI ×3 sim-to-real ×3 Citation Verification ×3 Token Budget ×3 budget allocation ×3 language modeling ×3 Mamba ×3 token compression ×3 Budget-Aware ×3 Context Distillation ×3 Open-source ×3 LLM Ranking ×3 causal inference ×3 foundation models ×3 Normalizing Flows ×3 evaluation metric ×3 retrieval evaluation ×3 LLM Recommendation ×3 Interoperability ×3 LLM Personalization ×3 data attribution ×3 Implicit Feedback ×3 Index Compression ×3 Game Theory ×3 Hierarchical RAG ×3 LLM Pre-training ×3 Captioning ×3 LLM Unlearning ×3 Role-Playing ×3 Item Tokenization ×3 self-evolution ×3 Text Embedding ×3 LLM Quantization ×3 multimodal reasoning ×3 self-evolving agents ×3 Knowledge distillation ×3 Coarse-to-Fine ×3 Contrastive Alignment ×3 encoder-decoder ×3 SIMD ×3 Student-Teacher ×3 GMM ×3 GPU ×3 Autoencoder ×3 Agentic Recommender Systems ×3 Policy Compliance ×3 adversarial robustness ×3 Agent Evolution ×3 Prefix Caching ×3 Multi-Agent Orchestration ×3 embedding ×3 skill retrieval ×3 Few-Step Generation ×3 Deep Search ×3 Audit ×3 Clarification ×3 Actor-Critic ×3 query routing ×3 Supervised Fine-Tuning ×3 Semantic Clustering ×3 Production System ×3 Kernel Optimization ×3 efficient retrieval ×3 Self-Training ×3 Peer Review ×3 Voice Agent ×3 Orchestration ×3 graph neural networks ×3 Question Decomposition ×3 bias mitigation ×3 Causal Discovery ×3 zero-shot ×3 sliding-window attention ×3 Data Mixing ×3 Sequential Modeling ×3 Multi-Reward RL ×3 Frozen Backbone ×3 MMR ×3 Environment Design ×3 Dense Reward ×3 fingerprinting ×3 Parameter Sharing ×3 Collective Intelligence ×3 cross-attention ×3 Retrieval Efficiency ×3 Parameter Efficiency ×3 text-to-video ×3 self-speculative decoding ×3 Temporal Retrieval ×3 Cost Reduction ×3 Latent Diffusion ×3 End-to-End Training ×3 Task Decomposition ×3 Prompting ×3 industrial deployment ×3 recall ×3 Gated Cross-Attention ×3 Human Alignment ×3 Agent Self-Evolution ×3 Replanning ×3 agentic evaluation ×3 Interactive Environments ×3 Long-Context Retrieval ×3 regularization ×3 evaluation metrics ×3 Mobile GUI Agent ×3 Audio-Video Generation ×3 Alignment Tax ×3 Layer Selection ×3 search ×3 Tool-Use Agents ×3 Recommender System ×3 Embedding Models ×3 long-horizon agent ×3 evaluation protocol ×3 Embedding Compression ×3 sponsored search ×3 Conversational Recommender Systems ×3 Prompt Tuning ×3 Logit Lens ×3 user simulation ×3 Semantic Matching ×3 Failure Recovery ×3 tool-use ×3 Conversational Memory ×3 Structured Output ×3 Agent Systems ×3 On-Policy RL ×3 Multi-Interest ×3 Multi-agent LLM ×3 softmax ×3 plug-and-play ×3 KB-VQA ×3 RLAIF ×3 linear-attention ×3 RL Fine-tuning ×3 structured pruning ×3 temporal reasoning ×3 Model Scaling ×3 Factuality ×3 Decision Transformer ×3 Knowledge Conflict ×3 Fine-Tuning ×3 parameter efficiency ×3 Shapley ×3 WebShop ×3 Budgeted Retrieval ×3 cross-modal retrieval ×3 user simulator ×3 Jailbreak ×3 Stateful Retrieval ×3 Retrieval Budget ×3 Decoupling ×3 Localization ×3 Permutation Invariance ×3 Extrapolation ×3 Regret ×3 Skill Composition ×3 Code Agent ×3 looped transformers ×3 Data Science ×3 long-horizon tasks ×3 Distribution Matching ×3 User Behavior ×3 Scientific Literature ×3 two-tower ×3 On-device LLM ×3 auditability ×3 conversational agents ×3 mathematical reasoning ×3 HCI ×3 Hierarchical Clustering ×3 candidate generation ×3 Data Efficiency ×3 diffusion language model ×3 masked diffusion ×3 Agent Learning ×3 Multi-Objective Ranking ×3 late-interaction ×3 Tool Synthesis ×3 Agent Optimization ×3 Qwen3 ×3 survey ×3 Reciprocal Rank Fusion ×3 risk control ×3 LLM Adaptation ×3 LLM Coding Agents ×3 LLM benchmark ×3 Visual Token Compression ×3 Multitask Learning ×3 Red-teaming ×3 Context Isolation ×3 Uplift Modeling ×3 measurement validity ×3 Co-evolution ×3 supervised fine-tuning ×3 autonomous agents ×3 load balancing ×3 Long-Context Training ×3 RL training ×3 tool retrieval ×3 LLM Benchmark ×3 generative search ×3 Experiential Learning ×3 Steering ×3 Triton Kernel ×3 Rubric-based RL ×3 evaluation benchmark ×3 task synthesis ×3 Agent Framework ×3 Lifecycle ×3 Idea Generation ×3 Bayesian Deep Learning ×3 Feature Selection ×3 differential privacy ×3 Token-Level Credit Assignment ×3 Cross-Tokenizer ×3 multi-turn dialogue ×3 Explainable AI ×3 Agent evaluation ×3 generative AI ×3 AI4AI ×3 co-evolution ×3 Cross-Domain Sequential Recommendation ×3 Mid-training ×3 provenance ×3 Generator-Evaluator ×3 Slate Recommendation ×3 Online Advertising ×3 open-ended generation ×3 semantic ID ×3 Training Data ×3 Small Language Models ×3 Environment Evolution ×3 Token Selection ×3 Continuous Diffusion ×3 Expert Parallelism ×3 program synthesis ×3 Multimodal Search ×3 Persistence ×3 recurrent memory ×3 Video World Model ×3 Modality Routing ×3 Open-World ×3 Prompt Compression ×3 deep research ×3 Task Synthesis ×3 mid-training ×3 music recommendation ×3 JSD ×3 web search ×3 Agentic Pipeline ×3 Neurosymbolic ×3 Generative Reward Model ×3 Machine Translation ×3 adaptive routing ×3 Transferability ×3 Agentic Design ×3 Scientific Agent ×3 Offloading ×3 Memory Consolidation ×3 DocID ×3 Feedback ×3 probabilistic forecasting ×3 Shared Memory ×3 measurement ×3 Throughput ×3 cross-model transfer ×3 Multi-turn Agents ×3 LightGCN ×3 Long context ×3 multimodal embeddings ×3 Adapter Composition ×3 black-box LLM ×3 causal intervention ×3 Document QA ×3 Post-Training Quantization ×3 query reformulation ×3 Length Extrapolation ×3 Agent Harness Optimization ×3 Recurrent Depth ×3 Task Vector ×3 Matryoshka Representation Learning ×3 representation geometry ×3 GSPO ×3 multi-agent LLM ×3 sycophancy ×3 multi-teacher distillation ×3 W4A4 ×3 reproducibility ×3 LLM auditing ×3 uncertainty ×3 Policy Enforcement ×3 reasoning traces ×3 Diffusion LLM ×3 Agent Infrastructure ×3 Expert Specialization ×3 Looped Models ×3 energy efficiency ×3 Game AI ×3 Decision Models ×3 Compute Allocation ×3 Agent Distillation ×3 Test-time scaling ×3 Trajectory Synthesis ×3 Decision Model ×3 KV cache quantization ×3 search agents ×3 Workflow Optimization ×3 stochastic rounding ×3 Sliding Window Attention ×3 hindsight ×3 RSI ×3 memory retrieval ×3 KL divergence ×3 Profiling ×2 scam detection ×2 agentic framework ×2 Autoregressive ×2 Few-Shot Learning ×2 Disentanglement ×2 agentic tasks ×2 Entropy Gate ×2 visual attention ×2 Auditability ×2 Olympiad ×2 disaggregated serving ×2 Spatial Understanding ×2 Autoregressive Generation ×2 Interactive Video ×2 CRS ×2 ModernBERT ×2 data-centric ×2 FalkorDB ×2 Physics Simulation ×2 LVLMs ×2 Memory Benchmark ×2 Memory-Augmented Agents ×2 Surrogate Model ×2 Web Search ×2 Camera Control ×2 Long Video Generation ×2 Massive Activations ×2 time series forecasting ×2 Feed-forward ×2 RAG Security ×2 Vector Database ×2 Zero-data ×2 Customer Service ×2 AI Governance ×2 Parallelization ×2 Multi-Policy ×2 low-resource ×2 Metric Learning ×2 Healthcare ×2 Repair ×2 Interaction ×2 reward learning ×2 hierarchical model ×2 power law ×2 Audio ×2 mixture density network ×2 Legal Reasoning ×2 Hierarchical Quantization ×2 score matching ×2 moment matching ×2 Semantic Features ×2 Outlier ×2 LLM security ×2 Inverse Propensity Weighting ×2 Evolutionary Algorithm ×2 EMA ×2 Image-to-Video ×2 stability ×2 generative model ×2 Audio LLM ×2 covariate shift ×2 Overthinking ×2 Multi-task ×2 Citation Faithfulness ×2 Embeddings ×2 Cost-Effectiveness ×2 Second-Order Optimization ×2 constrained decoding ×2 Retrieval Heads ×2 Lightweight Model ×2 latent representation ×2 User Interest Modeling ×2 Tool-Using Agents ×2 distributed training ×2 Test-Driven Development ×2 Generative Recommender ×2 Gradient Starvation ×2 symmetry ×2 SwiGLU ×2 Vision-Language Alignment ×2 streaming video ×2 interference ×2 Fashion ×2 Text-Image Generation ×2 I2I Retrieval ×2 Memory Construction ×2 creativity ×2 Trajectory Generation ×2 Upper Confidence Bound ×2 Exploration-Exploitation ×2 Agentic Inference ×2 FP4 Quantization ×2 personal assistant ×2 draft tree ×2 Citation ×2 Multi-Granularity ×2 TextGrad ×2 defense ×2 adversarial ×2 CFG ×2 Anomaly Detection ×2 Thompson sampling ×2 model collapse ×2 extrapolation ×2 Direct Preference Optimization ×2 Constrained Optimization ×2 Heavy-tailed ×2 Agent Trajectories ×2 Semantic Planning ×2 Editing ×2 episodic memory ×2 Representation Autoencoder ×2 Shampoo ×2 Monte Carlo ×2 Text-to-3D ×2 Omni-Modal ×2 Hyperparameter Transfer ×2 Scaffold Optimization ×2 Agent Debugging ×2 Hierarchical Skills ×2 Autonomous Agent ×2 Terminal ×2 sequence generation ×2 Local Deployment ×2 3D Grounding ×2 frontier models ×2 video benchmark ×2 Error Analysis ×2 Automated Theorem Proving ×2 VLN ×2 Self-awareness ×2 Generative Re-Ranking ×2 Token Diversity ×2 Greedy Decoding ×2 Skill Lifecycle ×2 Agent Protocol ×2 Few-shot ×2 convex optimization ×2 Large Reasoning Models ×2 representation steering ×2 SDE ×2 ODE ×2 DDIM ×2 Language Consistency ×2 multi-reward ×2 SNR ×2 Production Deployment ×2 Runtime Verification ×2 Multi-Hop QA ×2 Bilevel Optimization ×2 Post-hoc ×2 Data Ordering ×2 Temporal Knowledge ×2 Video-LLM ×2 Ads ×2 Graph Traversal ×2 model ranking ×2 Multi-Turn ×2 LLM Planner ×2 scaling law ×2 GraphMemory ×2 Unsupervised ×2 Visual Search ×2 Embedding Collapse ×2 Effective Rank ×2 sampled softmax ×2 Long-tail Recommendation ×2 Intent Tracking ×2 Skill ×2 Environment Co-evolution ×2 unified model ×2 Confidence-Aware ×2 gradient projection ×2 Generative UI ×2 Re-ranking ×2 Longitudinal Benchmark ×2 data generation ×2 multi-reward RL ×2 Pareto optimization ×2 GUI agents ×2 weight decay ×2 Search Ranking ×2 Graph Contrastive Learning ×2 Cross-modal Retrieval ×2 backdoor defense ×2 gradient analysis ×2 social simulation ×2 Dense Supervision ×2 Pointer Network ×2 Online Inference ×2 Full-duplex ×2 MoE-LoRA ×2 Divide-and-Conquer ×2 DeepResearch ×2 Query Routing ×2 LLM Augmentation ×2 LLM supervision ×2 value model ×2 MTP ×2 Qwen ×2 memory systems ×2 operations research ×2 Latent Compression ×2 peer review ×2 tree search ×2 Skill Internalization ×2 Value Alignment ×2 Recommendation System ×2 Multi-Task ×2 proximity graph ×2 index compression ×2 Connectivity ×2 Self-evolving Agents ×2 Cost-Quality Tradeoff ×2 influence function ×2 Cooperation ×2 Online Calibration ×2 heterogeneous retrieval ×2 agent benchmark ×2 state tracking ×2 Factual Grounding ×2 LSTM ×2 warm start ×2 scaling ×2 AndroidWorld ×2 Factual QA ×2 Program-of-Thought ×2 Iterative Optimization ×2 User Profile ×2 Medical QA ×2 gradient interference ×2 data mixture ×2 Representation Analysis ×2 Synthetic Benchmark ×2 Memory Evaluation ×2 Memory Lifecycle ×2 Watermarking ×2 Time-Series Anomaly Detection ×2 Whitening ×2 failure modes ×2 Defense ×2 CVR prediction ×2 Late-Interaction ×2 IO-Aware ×2 multimodal RAG ×2 Metropolis-Hastings ×2 Evaluation Bias ×2 counterfactual replay ×2 LLM adaptation ×2 transformer architecture ×2 representation analysis ×2 LLM Efficiency ×2 Incremental Update ×2 Dataset Distillation ×2 First-Order Optimization ×2 MDL ×2 Document Representation ×2 QA ×2 KL-regularized RL ×2 Convergence ×2 throughput ×2 vMF ×2 Critic Model ×2 weak-to-strong generalization ×2 Privacy-Utility Trade-off ×2 Novelty ×2 Deep Research Agents ×2 Multi-Agent Framework ×2 Multimodal Evaluation ×2 Financial Agents ×2 Governance ×2 quality estimation ×2 dual-tower ×2 self-evolving agent ×2 skill learning ×2 Autonomous Training ×2 Codebook Collapse ×2 Lean4 ×2 Human Evaluation ×2 Time-series Forecasting ×2 Real Estate ×2 SQL Generation ×2 autoencoder ×2 Advantage Weighting ×2 System Prompt ×2 Efficient Reasoning ×2 Non-Parametric ×2 LLM-Agent ×2 Cost Control ×2 Adaptive Sampling ×2 Auditable ×2 reasoning efficiency ×2 Metric Space ×2 Agent Skill ×2 Memory System ×2 Adam ×2 detection ×2 Experience Internalization ×2 Adaptive Planning ×2 LLM Ensemble ×2 Invariant Learning ×2 Regret Minimization ×2 Software Evolution ×2 Evidence Tracing ×2 context distillation ×2 retrieval-augmented generation ×2 document compression ×2 Error Propagation ×2 Imitation Learning ×2 LLM control ×2 LLM-reasoning ×2 Data Leakage ×2 MLE ×2 Scalable ×2 live streaming ×2 embodied AI ×2 Multi-Agent Reasoning ×2 Pipeline Parallelism ×2 KV-cache Reuse ×2 re-ranking ×2 User Embedding ×2 tool learning ×2 Data Analysis ×2 Autoregressive Decoding ×2 Negative Sampling ×2 Multiple Instance Learning ×2 few-shot ×2 associative memory ×2 View Consistency ×2 hidden state ×2 Language Model ×2 Autoregressive Models ×2 AI agent ×2 PageRank ×2 BPTT ×2 Alternating Optimization ×2 Forget-Retain Trade-off ×2 Multi-Hop Reasoning ×2 human evaluation ×2 autonomy ×2 benchmark construction ×2 Spectral Regularization ×2 MAP-Elites ×2 Long-term Conversation ×2 Hierarchical Graph Memory ×2 Tool-Augmented Reasoning ×2 mixture of experts ×2 W4A4KV4 ×2 Bidirectional Attention ×2 Real-Time ×2 process reward ×2 Attention Residuals ×2 Actionable Feedback ×2 Encoder-Decoder ×2 Adaptive Clipping ×2 VLM Agent ×2 KV Cache Reuse ×2 off-policy RL ×2 token-level supervision ×2 iOS ×2 Length Control ×2 large reasoning models ×2 LLM-RAG ×2 robust optimization ×2 Shared Context ×2 Preference Elicitation ×2 Multimodal Sequential Recommendation ×2 Image Tokenizer ×2 brand exposure ×2 Test-time Adaptation ×2 Multimodal QA ×2 Evidence Compression ×2 Hypothesis Generation ×2 video world model ×2 Real-Time Interaction ×2 Verifiability ×2 Multimodal Generation ×2 Parallel Sampling ×2 Rollout Allocation ×2 Information Seeking ×2 Behavior Prediction ×2 Medical Reasoning ×2 Indic Languages ×2 Audio Generation ×2 soft tokens ×2 Environment Engineering ×2 Unified Multimodal ×2 Interleaved Generation ×2 Asynchronous Inference ×2 Visual Prompting ×2 Multi-modal LLM ×2 Latent CoT ×2 Information Bottleneck ×2 agentic coding ×2 Poisoning Attack ×2 Click Model ×2 Query-Adaptive ×2 attention mechanism ×2 query expansion ×2 relevance assessment ×2 Workspace ×2 Multi-Query ×2 Self-Assessment ×2 Code-as-Action ×2 Speech LLM ×2 Subagent ×2 Influence Functions ×2 Model Efficiency ×2 root cause analysis ×2 Clustering ×2 GPU Acceleration ×2 time series ×2 pattern matching ×2 GUI Automation ×2 Second-order ×2 CNN ×2 adaptive stopping ×2 Standardization ×2 A2A ×2 Embedding Distillation ×2 Temporal Encoding ×2 query rewriting ×2 production deployment ×2 agent reliability ×2 sample complexity ×2 personal agent ×2 code review ×2 Multimodal Grounding ×2 Short-Video ×2 KnowledgeGraph ×2 Slot Attention ×2 Linear Complexity ×2 Long-horizon Generation ×2 Conflict Filtering ×2 Generative Reward Models ×2 Coverage ×2 pessimism ×2 agent trajectories ×2 multi-agent debate ×2 safety evaluation ×2 LLM Scaling ×2 semantic graph ×2 LLM Embedding ×2 Lookup-Free Quantization ×2 dLLM ×2 overthinking ×2 Memory Governance ×2 Agent Trajectory ×2 Active Perception ×2 Monte Carlo Tree Search ×2 generative retrieval ×2 failure analysis ×2 n-gram ×2 Agent Runtime ×2 Session State ×2 multi-turn tool use ×2 生成式推荐 ×2 Evidence Extraction ×2 Structured State ×2 Meta-learning ×2 SLERP ×2 Synthetic Query ×2 Factual Consistency ×2 Entropy Collapse ×2 Token-Level Reweighting ×2 Engram ×2 Retrieval-Augmented QA ×2 Fault Injection ×2 meta-evaluation ×2 adversarial retrieval ×2 admission control ×2 SGD ×2 Information Geometry ×2 Missing Not At Random ×2 overfitting ×2 Transparency ×2 Monitorability ×2 Streaming RAG ×2 Virtual Try-On ×2 semantic caching ×2 Agent Verification ×2 Runtime Monitoring ×2 4-bit quantization ×2 AMD GPU ×2 Annotation-Free ×2 LLM-agents ×2 Trajectory Supervision ×2 Jailbreak Defense ×2 instruction-following ×2 Pairwise Comparison ×2 backtracking ×2 Biomedical QA ×2 Spectral Collapse ×2 Bayesian Updating ×2 Enterprise Agent ×2 Retrieval Benchmark ×2 MLP ×2 context rot ×2 voice agent ×2 small language models ×2 Structure-Aware ×2 multimodal agents ×2 Attention Head ×2 Offline Evaluation ×2 agentic verification ×2 Agent Simulation ×2 Agent Skill Evolution ×2 parallel reasoning ×2 length generalization ×2 Experience Learning ×2 full-duplex ×2 agentic AI ×2 Bayesian inference ×2 Data Management ×2 Dependency Graph ×2 Industrial Scale ×2 Process Reward Model ×2 Group-based RL ×2 3D Reasoning ×2 Mode Collapse ×2 Static Analysis ×2 Code Agents ×2 classifier-free guidance ×2 BFCL ×2 lightweight ×2 Semantic Consistency ×2 LTT ×2 rating prediction ×2 LLM Fine-Tuning ×2 Top-k ×2 multimodal pretraining ×2 GPU kernel ×2 Unigram ×2 token pruning ×2 survey sampling ×2 Trust Boundary ×2 proper scoring rules ×2 dialogue evaluation ×2 forecasting ×2 LLM interpretability ×2 Geometry ×2 Representation Drift ×2 benchmark quality ×2 Industrial-scale ×2 Data-Centric ×2 ILP ×2 Constitutional AI ×2 bandit ×2 transferability ×2 reward-guided generation ×2 Agentic Abstention ×2 Adaptive RAG ×2 Selective Prediction ×2 mixture-of-experts ×2 Privilege Illusion ×2 Memory Agents ×2 On-Device ×2 ALFWorld ×2 Submodular Optimization ×2 Multi-teacher distillation ×2 adaptive policy ×2 Graph Navigation ×2 Multi-hop ×2 Hard Negative Sampling ×2 feedback-driven ×2 Capability Integration ×2 sequence parallelism ×2 One-Step Generation ×2 memory hierarchy ×2 token-level ×2 Human-Agent Collaboration ×2 Agentic Evaluation ×2 GPT-5 ×2 Reinforcement learning ×2 Long Document QA ×2 uncertainty calibration ×2 metacognition ×2 Behavior Cloning ×2 Hypothesis Testing ×2 prototype alignment ×2 Whisper ×2 Graph Reasoning ×2 Traceability ×2 Dual-Stream ×2 Knowledge Graph QA ×2 automated optimization ×2 emergence ×2 Self-Supervision ×2 Cost Model ×2 Gemma 4 ×2 Trie ×2 LLM-based Agents ×2 Batch Size ×2 Explainable Recommendation ×2 Neural Architecture Search ×2 Distractor Robustness ×2 AI coding agents ×2 Sudoku ×2 low-rank adaptation ×2 hierarchical RL ×2 Evidence Replay ×2 Granularity Control ×2 self-reflection ×2 out-of-distribution ×2 Filtered Search ×2 weight interpolation ×2 cross-modal transfer ×2 Embodied Agents ×2 Social Simulation ×2 Object-Centric ×2 Vision Encoder ×2 Experiential Memory ×2 reasoning effort ×2 vector steering ×2 inference-time control ×2 Product Search ×2 Push Notification ×2 Autoregressive Model ×2 Graph RAG ×2 Time-Aware ×2 privileged information ×2 Jensen-Shannon divergence ×2 Real-World Tasks ×2 Attention Calibration ×2 uplift modeling ×2 optimal stopping ×2 Throughput Optimization ×2 misinformation ×2 throughput optimization ×2 Retrieval Augmented Generation ×2 Agentic Post-Training ×2 Failure-Driven Evolution ×2 Memory Bank ×2 Wikipedia ×2 LLM reranker ×2 multi-hop retrieval ×2 inference-time adaptation ×2 Dataset Search ×2 vector compression ×2 lightweight fine-tuning ×2 Linear RNN ×2 Legal AI ×2 Structured Index ×2 Outcome Reward ×2 Trajectory Distillation ×2 Value Model ×2 Collaborative Agents ×2 Inductive Reasoning ×2 internal representations ×2 Length Generalization ×2 Training Framework ×2 softmax approximation ×2 Multi-Interest Retrieval ×2 structured generation ×2 Latency Reduction ×2 question answering ×2 Masked Diffusion Models ×2 Multimodal Understanding ×2 unsupervised ×2 Markov Chain ×2 memorization ×2 Conversational Retrieval ×2 Model Calibration ×2 SFT+RL ×2 Agent Orchestration ×2 Sensemaking ×2 Emergent Misalignment ×2 enterprise ×2 AgentOS ×2 Off-policy ×2 Automated ML ×2 learning rate schedule ×2 agentic inference ×2 Boolean Retrieval ×2 Regret Bound ×2 Weak-to-Strong Generalization ×2 Query Drift ×2 Memory Tokens ×2 Egocentric Video ×2 Multimodal Representation ×2 User Behavior Modeling ×2 Training Pipeline ×2 human feedback ×2 scaffolding ×2 Video QA ×2 Infrastructure ×2 Modular Design ×2 redundancy ×2 Unified Framework ×2 Evaluation Benchmark ×2 Budget Allocation ×2 Self-Evolving Memory ×2 Cross-Platform ×2 generalizability theory ×2 memory lifecycle ×2 modular architecture ×2 Composed Image Retrieval ×2 Residual Scaling ×2 Vision-Language-Action ×2 benchmark evaluation ×2 Advantage Shaping ×2 Stress Testing ×2 Graph Matching ×2 Experience Memory ×2 Offline-Online Decoupling ×2 Ad Ranking ×2 Trust Calibration ×2 small language model ×2 Multi-step Reasoning ×2 chess ×2 Context Parallelism ×2 Emergent Behavior ×2 Resource Constraints ×2 Capability Taxonomy ×2 Mixed-Initiative ×2 Risk Assessment ×2 Linear Probing ×2 AppWorld ×2 Workflow Synthesis ×2 Multi-turn RL ×2 Multi-domain ×2 Schema-based ×2 task decomposition ×2 Parallel Generation ×2 cross-lingual ×2 Memory management ×2 Wasserstein distance ×2 source attribution ×2 empirical study ×2 semantic representation ×2 Constrained Beam Search ×2 Uncertainty-Aware ×2 Dialogue ×2 embedding adaptation ×2 LLM-training ×2 LLM-as-a-Coach ×2 Attention Fusion ×2 asynchronous RL ×2 language models ×2 inference-time intervention ×2 Synthetic Data Generation ×2 Multi-Agent RAG ×2 Evidence-Grounded QA ×2 drift detection ×2 memory-efficient training ×2 customer service ×2 Heterogeneous LLM ×2 FFT ×2 query decomposition ×2 PLAID ×2 Topology Preservation ×2 document set selection ×2 Procedural Data ×2 evaluation methodology ×2 Multi-Domain ×2 BPR ×2 Statistical Inference ×2 Rank Allocation ×2 Ranking Models ×2 conversational recommender systems ×2 transparency ×2 Prolog ×2 Plug-and-play ×2 Cross-domain Recommendation ×2 Tool-Calling ×2 few-step generation ×2 MIPS ×2 Self-supervision ×2 Discrete Flow Matching ×2 EMA teacher ×2 Error Localization ×2 Probe ×2 Representation Engineering ×2 LLM decoding ×2 Language Models ×2 Meta-Skills ×2 contextual bandit ×2 delayed feedback ×2 cost-quality trade-off ×2 cross-platform ×2 multi-armed bandit ×2 query-conditioned retrieval ×2 Embedding Geometry ×2 filter bubble ×2 Agentic Planning ×2 decoder-only ×2 OmniLLM ×2 small model ×2 agent auditing ×2 Native Memory ×2 ProgramBench ×2 memory-augmented LLM ×2 cost-efficiency ×2 hybrid architecture ×2 Attention Mask ×2 Hierarchical Planning ×2 Finance ×2 Contextual Correction ×2 item embeddings ×2 Data Flywheel ×2 Video Captioning ×2 Image Grounding ×2 Visual Retrieval ×2 Teacher-Student Distillation ×2 Long-Sequence Modeling ×2 Task Construction ×2 Policy Discovery ×2 token-cost ×2 text generation ×2 Graph Index ×2 Industrial Recommender Systems ×2 Reasoning Models ×2 Numerical Precision ×2 Elo Rating ×2 Telemetry ×2 GPU Kernel Optimization ×2 prompt engineering ×2 Counterfactual Reasoning ×2 multimodal memory ×2 Trajectory Evaluation ×2 Human Feedback ×2 Black-box Attack ×2 Detection ×2 test-time optimization ×2 gradient attribution ×2 multimodal recommendation ×2 Structured Search ×2 DSL ×2 agentic workflow ×2 ALiBi ×2 passkey retrieval ×2 Bi-Encoder ×2 Enterprise Workflows ×2 Data Contamination ×2 Fused Loss ×2 Graph Topology ×2 Agentic Recommender ×2 RVQ ×2 Cycle Consistency ×2 shortcut learning ×2 Video-Language Models ×2 Research Agent ×2 A-B testing ×2 Token-level Credit Assignment ×2 harness optimization ×2 gender bias ×2 attention steering ×2 Conflict Resolution ×2 Document Parsing ×2 MMoE ×2 consistency ×2 Small LLM ×2 Agentic Intelligence ×2 Information Extraction ×2 watermarking ×2 Autonomous Driving ×2 LLM Router ×2 model cascade ×2 Inference optimization ×2 Selective Trust ×2 product recommendation ×2 Rubric Scoring ×2 Identifiability ×2 audit ×2 KV cache reuse ×2 Document Understanding ×2 Diffusion LLMs ×2 Genetic Algorithm ×2 guardrails ×2 Energy Efficiency ×2 decision-making ×2 long-horizon evaluation ×2 Memory Retrieval ×2 Modality Bias ×2 Preference Distillation ×2 End-to-end ×2 Representation Collapse ×2 Ranking Loss ×2 LLM quantization ×2 post-training quantization ×2 round-to-nearest ×2 low-bit inference ×2 Implicit Reward ×2 On-policy self-distillation ×2 Textual Gradient ×2 Knowledge Editing ×2 RL post-training ×2 GPU scheduling ×2 latency optimization ×2 Multi-modal Recommendation ×2 Scaffolding ×2 Item Relations ×2 Hybrid Linear Attention ×2 MTEB ×2 Self-Evolving Agent ×2 text embeddings ×2 Pareto frontier ×2 Cost-Accuracy Trade-off ×2 AI Scientist ×2 Online Evaluation ×2 Causal Distillation ×2 Falsification ×2 Tool use ×2 Enterprise agents ×2 Data Agent ×2 Process Metrics ×2 CVR ×2 Diffusion Drafter ×2 Brier score ×2 Jailbreak Robustness ×2 Evaluation Metric ×2 Hadamard ×2 Data Repetition ×2 Long-form Video ×2 Deep Learning ×2 LLaDA ×2 data diversity ×2 decision making ×2 Power Sampling ×2 Harness Scaling ×2 trustworthy AI ×2 LLM reliability ×2 frozen LLM ×2 Bayesian Optimization ×2 Fake News Detection ×2 Off-policy RL ×2 Evolution Strategies ×2 Indirect Prompt Injection ×2 Agent Skill Retrieval ×2 multi-agent RL ×2 Retrieval Augmentation ×2 Multi-agent RL ×2 self-feedback ×2 field experiment ×2 Version Control ×2 AI-for-AI ×2 Evidence Synthesis ×2 LLM ranking ×2 diversity collapse ×2 Generative Ranking ×2 precision ×2 Modularity ×2 Environment Generation ×2 Scientific Software ×2 contrastive explanation ×2 tabular data ×2 Credit assignment ×2 environment synthesis ×2 Skill Transfer ×2 Sequential Monte Carlo ×2 classification ×2 evidence grounding ×2 Graph Generation ×2 Topology Design ×2 cache eviction ×2 Code-as-Policy ×2 Dual-Agent ×2 Probabilistic Forecasting ×2 evaluation framework ×2 failure prediction ×2 User Study ×2 Benchmark Evaluation ×2 adaptive reasoning ×2 Muon Optimizer ×2 Learning Rate ×2 Uncertainty Reduction ×2 Token Mixing ×2 Decoder-only LLM ×2 long-tail ×2 Expressivity ×2 runtime monitoring ×2 competitive programming ×2 Game Agent ×2 Cross-lingual Alignment ×2 reasoning trace ×2 Agent harness ×2 GAE ×2 Causal Attribution ×2 Contrastive Decoding ×2 watch-time prediction ×2 entity resolution ×2 Test-Time Evolution ×2 document retrieval ×2 Session-aware ×2 Persona Modeling ×2 Corpus Navigation ×2 Trajectory Data ×2 Cold-start Recommendation ×2 LLM Judges ×2 Quality-Cost Tradeoff ×2 long-horizon reasoning ×2 Navigation ×2 Frequency Domain ×2 Differential Privacy ×2 Agentic Benchmark ×2 Mental Health ×2 Replay ×2 Block Drafting ×2 Inference-Time ×2 Recoverability ×2 Identification ×2 Approximate Nearest Neighbor ×2 Prediction-Powered Inference ×2 Hybrid Search ×2 Causal Analysis ×2 Unified Backbone ×2 LALM ×2 Vector Index ×2 verbalized confidence ×2 Parameter Efficient ×2 machine translation ×2 Trajectory Learning ×2 non-determinism ×2 Stability ×2 Advertising ×2 Online A-B ×2 prefill ×2 dynamic routing ×2 Hierarchical RL ×2 Positional Bias ×2 Evidence Retrieval ×2 Slate Generation ×2 Agentic Verification ×2 Physical Reasoning ×2 Speaker Diarization ×2 Speech ×2 GPU utilization ×2 Production Systems ×2 Controlled Study ×2 Environment Reconstruction ×2 factuality ×2 LLM reranking ×2 Quantization-Aware Training ×2 Test Generation ×2 Time-Series Forecasting ×2 Citation Graph ×2 MLA ×2 token budget ×2 Preference Learning ×2 Complementary Recommendation ×2 Automation ×2 Doc2Query ×2 Task Dependency Graph ×2 Mixture-of-Agents ×2 LLM explanations ×2 Relevance Feedback ×2 Transformer interpretability ×2 Black-box LLM ×2 Kalman filter ×2 Reward Sparsity ×2 Evidence Localization ×2 SGLang ×2 Long-sequence Modeling ×2 prompt tone ×2 Grounded Generation ×2 Legal NLP ×2 Compliance ×2 Text Rendering ×2 tool-use agents ×2 Agent Skill Optimization ×2 Triton kernel ×2 Low Latency ×2 efficient training ×2 Fisher information ×2 profiling ×2 Caching ×2 Visual Agent ×2 Research Agents ×2 Sparse Memory ×2 Digital Agents ×2 Ontology ×2 offloading ×2 tokenizer ×2 QPP ×2 Inference-time Intervention ×2 Scientific Agents ×2 Efficient Retrieval ×2 Personalized Retrieval ×2 VLM Agents ×2 Robot Learning ×2 Audio-Visual ×2 Environment Construction ×2 LLM Decoding ×2 length inflation ×2 monitoring ×2 efficient decoding ×2 Diffusion LM ×2 score centering ×2 training-inference mismatch ×2 EAGLE ×2 Workflow Guidance ×2 Search Advertising ×2 Progressive Training ×2 Agentic Workflows ×2 Transformer Compression ×2 Question Generation ×2 VQ-VAE ×2 Knowledge Update ×2 decoding ×2 Latent Representation ×2 prefill optimization ×2 Jev ×2 BF16 ×2 simulation ×2 length-extrapolation ×2 Forward KL ×2 RFT ×2 Hierarchical Agent ×2 agent security ×2 AI safety ×2 KL regularization ×2 intent drift ×2 Rank Fusion ×2 Multi-Agent Coordination ×2 low-precision training ×2 Admission Control ×2 forward KL ×2 reverse KL ×2 Few-step Generation ×2 multi-objective ×2 Teacher Adaptation ×2 Recurrent State ×2 AI Safety ×2 Tabular Foundation Model ×2 DiskANN ×2 Agentic Learning ×2 hard negatives ×2 INT4 ×2 Inference Engine ×2 Long-context LLM ×2 LLM efficiency ×2 Action Verification ×2 Code Execution ×2 Generative Engine Optimization ×2 Multi-Agent Collaboration ×2 Tool Creation ×2 Adaptive Granularity ×2 Cross-lingual ×2 Verifier ×2 cross-entropy ×2 query-agnostic compression ×2 Financial AI ×2 Personal Agent ×2 speculative execution ×2 LLM Embeddings ×2 Embedding Space ×2 embedding retrieval ×2 Semantic Compression ×2 context compaction ×2 Work Agents ×2 Visual Document Retrieval ×2 Experiment Selection ×2 frozen embeddings ×2 momentum ×2 Copy Mechanism ×2 System One ×2 STaRK ×2 Tool Evolution ×2 GAIA ×2 decision models ×2 Compositionality ×2 CUA ×2 multi-turn agents ×2 Skill Management ×2 Drug Discovery ×2 cost-accuracy tradeoff ×2 Autonomous Discovery ×2 High Recall ×2 Industrial A-B Test ×2 Graph Search ×2 Critic ×2 Agentic Loop ×2 Experience Replay ×2 dynamic programming ×2 LLM Search ×2 Multi-view Synthesis ×1 Control Decoupling ×1 Photorealism ×1 Smartphone ×1 Semantic MDP ×1 Skill Decomposition ×1 Graph ×1 Design Space ×1 Table Understanding ×1 Visual QA ×1 Real-world Images ×1 Structural Reasoning ×1 Long-CoT Reasoning ×1 Multi-Teacher Decoding ×1 Step-wise Collaboration ×1 Predictive Perplexity ×1 Hierarchical Attention ×1 Implicit Conflict ×1 CUPMEM ×1 continuous space ×1 per-token time ×1 conditional generation ×1 Representation Evaluation ×1 Axiomatic Framework ×1 Latent Thought ×1 Self-Generated Data ×1 streaming data ×1 cross-app reasoning ×1 early anticipation ×1 agentic ×1 preference alignment ×1 tool generation ×1 Video Diffusion ×1 edge-computing ×1 multi-condition ×1 diffusion-models ×1 condition-pruning ×1 adaptive-offloading ×1 audio-video generation ×1 cross-modal alignment ×1 speech-sound harmonization ×1 human-centric ×1 multimodal synthesis ×1 Generalized Category Discovery ×1 Relational Pattern Consistency ×1 Semi-supervised Learning ×1 Novel Class Discovery ×1 Soft Decomposition ×1 Pre-Task ×1 Synthetic Practice ×1 Intent Recognition ×1 Inconsistency ×1 Dual-Pathway ×1 Prototype Matching ×1 step-level reward ×1 multi-agent pathfinding ×1 decentralized planning ×1 communication ×1 Pixel-space Generation ×1 In-context Generation ×1 Scalable Foundation Model ×1 token-level aggregation ×1 Hölder mean ×1 Anti-Self-Distillation ×1 Pointwise Mutual Information ×1 Reasoning RL ×1 vocabulary hijacking ×1 kill chain ×1 progression modeling ×1 psychological manipulation ×1 Multimodal Fact-Checking ×1 LLM-assisted Annotation ×1 Visual Misinformation ×1 Teacher Exposure ×1 Adversarial Optimization ×1 Policy Bank ×1 GALA Benchmark ×1 Delta Attention ×1 Selective Routing ×1 Hodge Decomposition ×1 Simplicial Complex ×1 Topological Data Analysis ×1 adaptive compression ×1 bandit control ×1 particle filter ×1 MiniGrid ×1 VLMs ×1 Soft Ranking ×1 Meta-cognition ×1 Knowing-Doing Gap ×1 Representation Probing ×1 360 Panorama ×1 Spherical Cross-Attention ×1 Pano-native ×1 Equirectangular Projection ×1 Image Super-Resolution ×1 Text Image ×1 Uncertainty Modeling ×1 Single-Step Diffusion ×1 Visual Shortcut ×1 Self-distillation ×1 Execution ×1 Industrial ×1 Adaptive Coordination ×1 Path Diversity ×1 Unified Models ×1 Functional Token ×1 Next-Token Prediction ×1 LA-GRPO ×1 Sparsity ×1 Binary Mask ×1 Causal Forcing ×1 Few-Step Sampling ×1 Evolutionary Merging ×1 Discrete Latent Codes ×1 Steering Vectors ×1 Low-data Fine-tuning ×1 spherical flow matching ×1 latent space geometry ×1 geodesic paths ×1 ReDial ×1 Avatar Reconstruction ×1 Feed-forward Network ×1 FLAME ×1 Real-time Animation ×1 Open-Ended ×1 Test-time adaptation ×1 layered design ×1 graphic design ×1 MSSP ×1 DMFT ×1 Innovation Methodologies ×1 Convergence Detection ×1 Patent Generation ×1 Mean Shift ×1 Interacting Particle Systems ×1 Quadrature ×1 Gradient Flow ×1 lottery ticket hypothesis ×1 sparse subnetwork ×1 quantum-inspired ×1 dequantization ×1 ridgelet transform ×1 Motion Reward ×1 Human Motion ×1 Hierarchical Models ×1 Broadcast Process ×1 Context Length ×1 Dynamics-Blindness ×1 Action Chunking ×1 Multi-task RL ×1 Online Distillation ×1 Mean-Matching ×1 Guardrail ×1 Policy Induction ×1 Visual Evidence ×1 End-to-end RL ×1 Hierarchical rollout ×1 Stagewise co-evolution ×1 Meta-agent ×1 Discriminative Utility Estimation ×1 DueDiligence ×1 HallucinationMitigation ×1 Intelligent Tutoring System ×1 Algorithm Tracing ×1 Socratic Method ×1 Simulated Student ×1 Profile Generation ×1 Evidence Ecosystem ×1 GUI Navigation ×1 Prospection ×1 Content Freshness ×1 Expiration Prediction ×1 Risk-Sensitive RL ×1 Entropic Risk ×1 Best Policy Identification ×1 Concentration Bounds ×1 Semantic Control ×1 Feature Interpretation ×1 Activation Sparsity ×1 Agentic Evolution ×1 Knowledge Network ×1 REINFORCE ×1 Competitive Programming ×1 Patch Repair ×1 tabular foundation model ×1 image retrieval ×1 context-aware search ×1 visual-semantic embedding ×1 abstraction levels ×1 persistent homology ×1 spectral sheaf ×1 cold start ×1 seasonal spikes ×1 foundation model adapter ×1 3D Scene Editing ×1 Residual Field ×1 Text-conditioned ×1 Multi-view Consistency ×1 DeltaScene Dataset ×1 multimodal deep research ×1 source visual evidence ×1 report generation ×1 evidence verification ×1 Steganography ×1 Embedding Integrity ×1 Provenance Defense ×1 Aggregation ×1 Field Experiment ×1 Emotional Escalation ×1 Tracing ×1 Span-level ×1 TRAIL benchmark ×1 Linguistic Adaptation ×1 Hawthorne Effect ×1 Distributed System ×1 URDF ×1 Articulated Objects ×1 Dataflow ×1 System ×1 embeddings ×1 matryoshka ×1 3D-ML ×1 strategy-nudging ×1 math-reasoning ×1 LLM Tutors ×1 Educational Safety ×1 Reasoning Paradox ×1 GUI Critic ×1 Affordance ×1 Multi-role ×1 Value Estimation ×1 End-to-End Planning ×1 Proposal Selection ×1 Element-level Grounding ×1 Open-vocabulary Detection ×1 Pluralistic Alignment ×1 vision-language judge ×1 feature learning ×1 spectral algorithm ×1 random matrix theory ×1 Clinical Reasoning ×1 Probabilistic Scoring ×1 Beta-Binomial ×1 lower bound ×1 Multi-Shot Video Generation ×1 Entity Consistency ×1 Visual Narrative ×1 Graph-Constrained Generation ×1 Verifier Agent ×1 IRAC ×1 garment customization ×1 autoregressive model ×1 interactive ×1 SpeculativeDecoding ×1 LatencyModel ×1 Little'sLaw ×1 Isometry ×1 Grep ×1 Hybrid Action Spaces ×1 Mixed Gradient Estimator ×1 Pathwise Gradients ×1 Score-Function ×1 Proximal Policy Optimization ×1 CBR ×1 Schema-Faithful Execution ×1 Langevin dynamics ×1 generative sampling ×1 Personalized AI ×1 Image Aesthetics ×1 LLM Interviews ×1 Human-AI Comparison ×1 Long-Horizon Animation ×1 Latent Flow ×1 Drift Mitigation ×1 HS Tariff Classification ×1 Rule Reasoning ×1 Benchmark Audit ×1 Browser Fingerprinting ×1 Behavioral Trace ×1 Side-Channel Attack ×1 Window-Level Optimization ×1 Drafter Training ×1 Self-Improving AI ×1 Transformer-Mamba Hybrid ×1 Neural Architecture Design ×1 Scientific Contribution Graph ×1 Technological Roadmapping ×1 Prerequisite Prediction ×1 AI-NLP ×1 Automated Discovery ×1 AWQ ×1 Video Anomaly Detection ×1 Test-Time ×1 Compositional Generation ×1 Inference-Time Guidance ×1 Classifier Guidance ×1 Continuous Treatment ×1 Individual Treatment Effect ×1 Circuit Attribution ×1 Premature Closure ×1 Medical LLMs ×1 Knowledge Integration ×1 backdoor attack ×1 triggerless backdoor ×1 Logging Policy ×1 Reward-Coverage Tradeoff ×1 Experimental Design ×1 Multi-agent Debate ×1 Argumentation ×1 Multimedia Verification ×1 A-QBAF ×1 Uncertainty-aware ×1 ProcedureQA ×1 Test-time compute ×1 Codeforces ×1 Point-Precise Control ×1 Semantic-Execution Gap ×1 Geometric Construction ×1 Mirror Descent ×1 Convex Duality ×1 Negative Entropy ×1 Emotion-Aware ×1 Discriminative Training ×1 Unified Retrieval-Ranking ×1 VAE Decoder ×1 Conditional Decoding ×1 Reference Attention ×1 Visual Quality ×1 SHAP ×1 feature attribution ×1 distribution estimation ×1 feature selection ×1 inventory control ×1 censored demand ×1 Bayesian regret ×1 normalizing flow ×1 Bayesian Termination ×1 mHealth ×1 JITAI ×1 Context-aware ×1 Walking ×1 Speaker Verification ×1 Hierarchical Tokenizer ×1 Decision Composition ×1 Gradient Orthogonality ×1 Backward Transfer ×1 importance weighting ×1 loss penalty ×1 empirical risk minimization ×1 Variational Policy Distillation ×1 EM ×1 Language Feedback ×1 Self-Recall ×1 3D Reconstruction ×1 Register Attention ×1 Ablation Studies ×1 Timeline Reconstruction ×1 Clinical NLP ×1 Future ×1 Concurrency ×1 Outlier Separation ×1 ECC ×1 Environment Understanding ×1 Explore-then-Act ×1 Educational Assessment ×1 Bloom's Taxonomy ×1 Model-based Evaluation ×1 Visual RAG ×1 Embedding Aggregation ×1 Financial Documents ×1 Semantic Shift ×1 Patch Tokens ×1 mental health ×1 temporal trajectory ×1 change point detection ×1 LLM Priors ×1 EM Algorithm ×1 PRP ×1 Noisy Pairwise Comparisons ×1 HIP ×1 Attention Dispersion ×1 Differential Attention ×1 Dynamic Graph Transformers ×1 Temporal Distribution Shift ×1 Continuous-Time Dynamic Graphs ×1 Deliberation Cascade ×1 Multimodal Training ×1 Modality Competition ×1 Variance Correction ×1 SOAP ×1 Large Batch ×1 Spectral Shaping ×1 Dynamic Schedule ×1 Population-Based Training ×1 Prompt-Only Learning ×1 CybORG CAGE-2 ×1 Probabilistic Masking ×1 cross-modal ×1 adversarial perturbation ×1 Schrödinger bridge ×1 entropy rate ×1 discretization ×1 low-NFE inference ×1 LLM-based-ad-auction ×1 Two-stage-ranking ×1 Logit-scoring ×1 Incentive-compatibility ×1 Generative-chatbot ×1 Sharpness-Aware Minimization ×1 Preconditioner ×1 Loss Geometry ×1 LLQR ×1 Temporal Indexing ×1 Write-Efficient ×1 MemTree ×1 generative AI agent ×1 transformer forecaster ×1 CO2 analytics ×1 simulated bifurcation ×1 manifold disentanglement ×1 pseudo-label ×1 foundation model distillation ×1 noise transition matrix ×1 knowledge injection ×1 distribution alignment ×1 N-gram Memory ×1 Cosine Gate ×1 prediction intervals ×1 local skewness ×1 asymmetric intervals ×1 Inference Speedup ×1 Image Forgery Detection ×1 few-shot anomaly detection ×1 residual alignment ×1 cross-category generalization ×1 DeliveryLog ×1 Optimistic Concurrency Control ×1 Observable-Read Isolation ×1 HTTP Middleware ×1 posterior collapse ×1 entropic autoencoder ×1 free energy ×1 Long-term Behavior ×1 Multimodal Benchmark ×1 Closed-Loop Verification ×1 Agentic Tool Invocation ×1 Instance-level Localization ×1 human attention modeling ×1 behavioral sequence learning ×1 implicit relevance signal ×1 digital twin signature ×1 agentic pipeline ×1 LTL ×1 Temporal reasoning ×1 Heterogeneous Features ×1 Data Imputation ×1 Message Passing ×1 second-order optimization ×1 runtime system ×1 asynchronous computing ×1 state offloading ×1 Economic Alignment ×1 REINFORCE++ ×1 Market Crash ×1 Sybil Attack ×1 Web Applications ×1 Brand Alignment ×1 Compound AI System ×1 Adversarial QC ×1 Skeleton-Founded Tokenization ×1 Asymmetric Unlikelihood Optimization ×1 SaaS benchmark ×1 DAG evaluation ×1 multi-component systems ×1 enterprise software ×1 Video Remaking ×1 Dual-Bridge Consistency ×1 Identity Consistency ×1 Verification Agent ×1 Code Synthesis ×1 Blender ×1 equivariance ×1 Zero-Expert ×1 Contextual Integrity ×1 Product-of-Experts ×1 Privacy Alignment ×1 Reverse KL Divergence ×1 Tool-Using Agent ×1 Underspecification ×1 AgentEdit-Bench ×1 proxy metrics ×1 top-k accuracy ×1 expert token rank ×1 Fine-grained Understanding ×1 Referring Expression ×1 Program Functions ×1 ReAct Agent ×1 Layer Parallelism ×1 Newton's Method ×1 SNLP ×1 IDN ×1 HCN ×1 Training Regularization ×1 Memory Control ×1 proactive ×1 memory-augmented agents ×1 multi-target aggregation ×1 lookback ×1 Review-Augmented ×1 Dual-Diffusion ×1 Lifecycle Governance ×1 MLRM ×1 Q-Former ×1 Semantic Generative Tuning ×1 Unified Multimodal Models ×1 Image Segmentation ×1 Generative Proxy ×1 Self-Healing ×1 Contrastive Evidence ×1 Expert Review ×1 TF-IDF ×1 GPT-4 ×1 evidence-seeking ×1 clinical reasoning ×1 On-Device RAG ×1 User Preference ×1 Query Steering ×1 Data-Bound Scaling ×1 Model-Aware ×1 CodeAgent ×1 HarnessEngineering ×1 ExecutableReasoning ×1 AgentInfrastructure ×1 α-entmax ×1 differentiable routing ×1 hierarchical attention ×1 Data Preparation ×1 Distributional Alignment ×1 semantic trajectories ×1 discussion structure ×1 human-AI comparison ×1 Topology-aware Sampling ×1 Multitask ×1 Reward Normalization ×1 Capability Data ×1 Watch-Time Prediction ×1 Long-Tailed Regression ×1 Multiplicative Correction ×1 Duration Bias ×1 Second-Stage Correction ×1 Sparse Rewards ×1 Table QA ×1 popularity prediction ×1 micro-video ×1 open-web context ×1 online adaptation ×1 SHORTS-CAST ×1 evidence-card ×1 Agent Generalization ×1 Programmatic Generators ×1 Cross-Environment Adaptation ×1 Expected Prediction Error ×1 Bayesian Prior ×1 Low-Active Users ×1 New Item Growth ×1 Phase-Aware Quantization ×1 Prefill-Decode Disaggregation ×1 ID-Free ×1 Modality-Aware PE ×1 Chebyshev Scalarization ×1 programmatic verification ×1 self-evolving verifier ×1 desktop task synthesis ×1 LLM-as-judge vs hard-coded ×1 INT2 ×1 proactive agents ×1 hidden intents ×1 Parallel Tool Calling ×1 Tool Prior Paradox ×1 Context Map ×1 Orientation Cache ×1 Cache Policy ×1 prompt refinement ×1 offline testing ×1 Token-Level Loss ×1 Search-Augmented ×1 hybrid tree ×1 Cross-Modal Synergy ×1 Saliency Masking ×1 user thoughts ×1 behavior prediction ×1 IMMR ×1 Text-guided Grounding ×1 E-commerce Retrieval ×1 Mosaic Augmentation ×1 Fine-grained Representation ×1 Multi-dimensional Preference ×1 General Preference Model ×1 LLM Wiki ×1 Knowledge Synthesis ×1 Cost Comparison ×1 Multi-document QA ×1 LLM optimization ×1 text artifact ×1 agent architecture ×1 side information ×1 Pareto search ×1 spectral whitening ×1 high-pass NS ×1 Lifelong Memory ×1 Profile Synthesis ×1 Rule Learning ×1 poisoning ×1 Graph Fraud Detection ×1 Semantic Camouflage ×1 One-class Learning ×1 Text-Attributed Graph ×1 Discriminator View ×1 NegativeSampling ×1 SequentialRecommendation ×1 HardNegativeMining ×1 KnowledgeDistillation ×1 MultiSourceScoring ×1 DivergenceReranking ×1 Tweedie Matching ×1 Manifold Constraint ×1 Guidance ×1 Probability Conservation ×1 Image Generation Agent ×1 Visual Experience Distillation ×1 Tool-Orchestrated ×1 temporal semantic cache ×1 MCP workflow optimization ×1 plan-execute ×1 industrial agent ×1 cache failure mode ×1 spectral bandits ×1 graph smoothness ×1 effective dimension ×1 interactive learning ×1 directed graph ×1 statistical inference ×1 theoretical guarantee ×1 Evolutionary Coding ×1 Search Analysis ×1 Edit Types ×1 checkpoints ×1 Generative Memory ×1 Decision-Content Decoupling ×1 goal-level attribution ×1 indirect influence ×1 contribution measurement ×1 interaction design ×1 agent initiative ×1 Simulation-Ready ×1 Physical Assets ×1 Physics Faithfulness ×1 equilibrium reasoners ×1 attractor landscape ×1 weight-tied models ×1 iterative reasoning ×1 adaptive computation ×1 constraint-driven ×1 Neuro-Symbolic Retrieval ×1 Automated Research ×1 Graph Reranking ×1 System-Level Programming ×1 Test Suite ×1 Loss-Adaptive ×1 Knowledge Acquisition ×1 reflective tuning ×1 Subsidy Allocation ×1 Ride-hailing ×1 Multi-city pretraining ×1 Log-transform ×1 Tail annealing ×1 Generative model ×1 Hill estimator ×1 MRCR ×1 GraphWalks ×1 SA-3D RoPE ×1 dual memory ×1 semantic memory ×1 MOIM ×1 application layer ×1 Memory-Efficient Fine-Tuning ×1 Chunk-wise Optimization ×1 Byte-balanced Partitioning ×1 Sparse Gradient ×1 Full-Parameter Training ×1 Detail-Condensing Queries ×1 Reconstruction-Generation Trade-off ×1 Feature Condensation ×1 Latent Diffusion Model ×1 Multi-step Derivation ×1 Source Provenance ×1 dimensionality reduction ×1 triplet loss ×1 Personality Perception ×1 Memory Update ×1 Linear Recurrent ×1 Preconditioned Optimizer ×1 Sophia ×1 Language Model Pretraining ×1 Score Distillation ×1 pairwise advantage ×1 lambda policy optimization ×1 ROUGE-L ×1 Latent Space Reasoning ×1 Audio-Visual Understanding ×1 Temporal Alignment ×1 Feature Supervision ×1 Embedding Learning Rate ×1 Trace Diagnostics ×1 Corpus-Level Analysis ×1 Model-Skill Orchestration ×1 Data Visualization ×1 SciVis ×1 Memory conflict ×1 JIT Compilation ×1 offline policy evaluation ×1 reserve price ×1 support-aware ×1 statistical certification ×1 ad marketplace ×1 replay methods ×1 Preference Weighting ×1 Efficient Fine-Tuning ×1 Safety Repair ×1 representation space ×1 DINOv2 ×1 Subproblem ×1 Visual Degradation ×1 MLLM Benchmark ×1 Spreadsheet Agent ×1 Domain-Specific ×1 Microsoft Excel ×1 Video Frame Sampling ×1 Temporal Surprise ×1 Taylor Series ×1 Long-Form Video QA ×1 Real-World Evaluation ×1 Self-Regulation ×1 Simulative Planning ×1 route generation ×1 public transit ×1 end-to-end ×1 Short-Drama Generation ×1 Spatial Consistency ×1 Narrative Pacing ×1 scientific forecasting ×1 temporal prediction ×1 Label Generation ×1 Document Ranking ×1 spatio-temporal reasoning ×1 MLLM evaluation ×1 active synthesis ×1 multi-agent pipeline ×1 Multi-Level Feedback ×1 Dynamic Taxonomies ×1 Observability ×1 Formal Proof ×1 DeepMind ×1 Instruction-Tuning ×1 Biomedical NLP ×1 Open-source LLM ×1 Clipping ×1 NSR ×1 Stochastic Rescue ×1 lexical semantics ×1 scene abstraction ×1 few-shot prompting ×1 structured representation ×1 word meaning in context ×1 Hyperfitting ×1 Rank Reordering ×1 Late-Stage LoRA ×1 Utility Evaluation ×1 Checkpoint-Restore ×1 OverlayFS ×1 Process Memory ×1 CoW ×1 Role Drift ×1 Policy Sharing ×1 Workflow Topology ×1 League Training ×1 Quadrotor Racing ×1 Fashion Image Retrieval ×1 Spherical Interpolation ×1 Gradient-Guided Sampling ×1 MultiAgent Coordination ×1 Economic Primitives ×1 Auditable Provenance ×1 Graph-Native Architecture ×1 Log Anomaly Detection ×1 Message-level ×1 Lightweight Router ×1 FastAPI ×1 Streaming API ×1 Boilerplate Reduction ×1 State Distribution ×1 experience reuse ×1 procedural abstraction ×1 Shapley interactions ×1 Banzhaf interactions ×1 proxy model ×1 tree ensembles ×1 residual correction ×1 tokenization ×1 integer programming ×1 subword ×1 Rényi efficiency ×1 tokenisation ×1 linear programming ×1 Hybrid Decoding ×1 Semantic Triples ×1 critique vector ×1 Uniform Diffusion ×1 Leave-One-Out ×1 demand elasticity ×1 neural demand model ×1 spline basis ×1 integrability ×1 multiproduct pricing ×1 DDPM ×1 Hint Guidance ×1 AMEL ×1 context accumulation ×1 negativity asymmetry ×1 evaluation robustness ×1 RLIF ×1 GDPO ×1 KL-Cov ×1 Noisy Channel ×1 Shannon Theory ×1 Non-monotonic Behavior ×1 Source-Level Rewriting ×1 Container Swapping ×1 Motion-Controlled Video Generation ×1 Causal Interaction ×1 Training-Free VLM ×1 Confidence-Aware Control ×1 MotiBench ×1 3D Vision ×1 Closed-loop ×1 Aesthetic Reasoning ×1 Visual Critique ×1 matching principle ×1 nuisance robustness ×1 Jacobian penalty ×1 chatbot evaluation ×1 retrieval bias ×1 false premise ×1 accuracy paradox ×1 System 1 and System 2 ×1 Reflective Thinking ×1 Adaptation ×1 Search-Augmented Reasoning ×1 OFSD ×1 Multi-Concept ×1 Sparse Disentanglement ×1 Text-space Optimization ×1 Validation Gate ×1 Rejected-Edit Buffer ×1 unified audio-language model ×1 real-time interaction ×1 Factual Freshness ×1 low-rank subspace ×1 KV cache projection ×1 capability selection ×1 no external signal ×1 motion-direction-blindness ×1 diagnosis ×1 instruction-tuning ×1 projector-loss ×1 concept-vector ×1 Predictability ×1 VPO ×1 reward diversity ×1 test-time search ×1 multi-objective RL ×1 multi-answer generation ×1 Candidate Calibration ×1 Behavior Evidence ×1 Frequency Decomposition ×1 BRIDGE ×1 instruction embedding ×1 benchmark robustness ×1 validity ×1 fidelity ×1 Adversarial Attacks ×1 AI Act ×1 Rubric-based Reward ×1 Automated Data Generation ×1 Approximate Attention ×1 I-O Complexity ×1 Lower Bound ×1 Keyframe Retrieval ×1 Step-level Guidance ×1 hyperparameter transfer ×1 PreferenceDisentanglement ×1 RewardDecoupling ×1 eCommerce ×1 Unit Test Generation ×1 Execution Matrix ×1 High-Resolution ×1 Semantic Patching ×1 Bottom-Up Search ×1 auxiliary loss ×1 environment prediction ×1 multilingual embeddings ×1 RankElastor ×1 Dense Scaling ×1 Noisy Embeddings ×1 Instruction Finetuning ×1 SymNoise ×1 NEFTune ×1 Curvature Regularization ×1 LLaMA ×1 Physics-informed ×1 Constitutive modeling ×1 Automated model discovery ×1 semantic towers ×1 generative recommenders ×1 Generative Semantic IDs ×1 Asymmetric Knowledge Transfer ×1 Next-Query Prediction ×1 Recursive Memory ×1 Multi-Turn Dialogue ×1 Augmentation ×1 State-Adaptive Memory ×1 OAT-GRPO ×1 Episodic Recall ×1 iterative alignment ×1 trajectory ×1 noise robustness ×1 multi-task ranking ×1 LLM finetuning ×1 user story ×1 Inference-Time Recurrence ×1 Damped Sub-steps ×1 Calibrated Belief ×1 Controlled Communication ×1 Long-Horizon Inference ×1 optimization geometry ×1 directional alignment ×1 Adaptive Depth ×1 Chance-Corrected Metric ×1 Bits-over-Random ×1 Personal Agents ×1 A2UI-Bench ×1 Next Implicit Token Prediction ×1 representation degeneration ×1 self-supervised target ×1 pre-training objective ×1 Spectral Retrieval ×1 Multi-Scale Sinc Convolution ×1 Multi-Agent Retrieval ×1 task-oriented dialogue ×1 streaming media ×1 dialogue state tracking ×1 persona-driven generation ×1 cascade router ×1 recency bias ×1 BM25 acceleration ×1 GPU retrieval ×1 Agent Aging ×1 Lifespan Engineering ×1 Memory Degradation ×1 Counterfactual Diagnosis ×1 long-horizon event stream ×1 cross-service coordination ×1 multi-device interaction ×1 Multi-Concept Personalization ×1 variance-adaptive ×1 Personalized Memory ×1 Storage Gating ×1 Memory Retention ×1 corruption robustness ×1 Semantic Caching ×1 Prompt Segmentation ×1 Cache Hit Rate ×1 Feature Sparsity ×1 Multi-Path Exploration ×1 Generative CTR ×1 Gradient Balancing ×1 Heterogeneity ×1 Self-Balancing Loss ×1 Neural Scaling ×1 Predictive Modeling ×1 Large-scale Training ×1 Functional Form ×1 fast weights ×1 depth-recurrence ×1 hybrid model ×1 missing modality ×1 candidate reranking ×1 evidence routing ×1 scale-vector ×1 preconditioning ×1 reparameterization ×1 Token Teachability ×1 Selective Distillation ×1 industrial AI ×1 execution signals ×1 system scaling ×1 context governance ×1 memory trust ×1 Generative Search ×1 Query-Conditioned ×1 Linear HSTU ×1 HFG-Attention ×1 Bias Amplification ×1 Legal Agent ×1 Hybrid RAG+Search ×1 Time-Aware Retrieval ×1 Multi-Behavior Recommendation ×1 Graph Information Bottleneck ×1 Audio Retrieval ×1 MMTEB ×1 MRL ×1 spectral entropy ×1 sample filtering ×1 Tool Call Reduction ×1 Dual-path rollouts ×1 RelevanceJudgment ×1 HierarchicalReasoning ×1 PreferenceOptimization ×1 Sequential CTR ×1 Latent-Query ×1 Target-Conditioned ×1 Query-Specific ×1 Density-driven ×1 Two-Stage Ranking ×1 contextual integrity ×1 PII leakage ×1 Lookahead Planning ×1 Cumulative Regression ×1 Training Data Distribution ×1 Mitigation ×1 Self-evolution ×1 AppDev ×1 Multi-level Memory ×1 Cross-agent Transfer ×1 SkillsBench ×1 Omnimodal LLMs ×1 Streaming Interaction ×1 Real-time Evaluation ×1 skill_distillation ×1 token_efficiency ×1 web_agent ×1 online_learning ×1 VisualWebArena ×1 cache_aware_prompting ×1 Failure-Correction ×1 Prediction Markets ×1 Lifelong Sequence ×1 Target Attention ×1 Source Monitoring ×1 Typed Memory Atoms ×1 Provenance-Scoped ×1 CognitiveGraph ×1 MentalModel ×1 RegulatoryLoops ×1 SelfRegulation ×1 User Preference Modeling ×1 textual gradients ×1 gradient dilution ×1 Parallel Thinking ×1 Information Sharing ×1 Collaborative Reasoning ×1 adjoint matching ×1 dual descent ×1 Signal Collapse ×1 Medical RAG ×1 Idle-Time Compute ×1 Future-State Prediction ×1 ProActEval ×1 ID Collision ×1 Zero-Collision Reassignment ×1 Video Production ×1 Harness Design ×1 Decentralized ×1 Parallel Experiments ×1 tool-call resampling ×1 thinking-acting gap ×1 Batchwise Information Sharing ×1 Weakness-aware Data Generation ×1 Distributed ×1 Heterogeneous Ensembles ×1 Asynchronous Communication ×1 Remote Sensing QA ×1 Traceable Reasoning ×1 Coverage Illusion ×1 Retrieval Cascade ×1 Production RAG ×1 multi-task learning ×1 relevance ranking ×1 ordinal relevance ×1 coefficient calibration ×1 log-probability proxy ×1 Parameter-Level Adaptation ×1 Dual-View ×1 Personalized MoE ×1 programming model ×1 type safety ×1 code-as-action ×1 Lost-in-Conversation ×1 history cleaning ×1 error tracing ×1 execution graph ×1 Finance LLM Agent ×1 Safety Harness ×1 Tool Monitoring ×1 Cascade Judge ×1 Inline Lifecycle ×1 MusicLLM ×1 Multi-stage Pre-training ×1 Next-behavior Prediction ×1 workspace ×1 Confidence Head ×1 Local Linear Attention ×1 Architecture-Optimizer Codesign ×1 Hardware-Aware ×1 LLM reviewers ×1 multi-dimensional evaluation ×1 retrieval-augmented verification ×1 argument mining ×1 Proactive Recommendation ×1 Reward Centering ×1 PRS ×1 Math ×1 Research-level ×1 α-Rényi ×1 Pointwise Evaluation ×1 Non-verifiable Domains ×1 bidirectional search ×1 training data generation ×1 Skill-based RL ×1 Difficulty-Aware Router ×1 Traffic Allocation ×1 Blending ×1 Linear Boosting ×1 VibeSearch ×1 Knowledge Graph Evaluation ×1 Asynchronous Tool Calling ×1 causal representation learning ×1 exponential decay ×1 RAT+ ×1 Noise Injection ×1 range filtering ×1 segment tree ×1 Logical Retrieval ×1 Query2Conf ×1 Retrieval Agents ×1 Configuration Optimization ×1 Workload-Specific Characteristics ×1 streaming video generation ×1 adaptive anchors ×1 autoregressive diffusion ×1 relative time encoding ×1 hidden state recurrence ×1 lightweight model ×1 guardrail ×1 Future Supervision ×1 causal discovery ×1 interactive reasoning ×1 structural causal models ×1 3D spatial reasoning ×1 geometric prior ×1 correspondence learning ×1 depth consistency ×1 self-supervised ×1 representational accuracy ×1 behavioral specification ×1 AI personalization ×1 interpretive layer ×1 BGE ×1 Contrastive Reflection ×1 Insight Discovery ×1 Non-parametric Learning ×1 Hybrid MAS ×1 Cloud-Edge Inference ×1 Task-dependent Architecture ×1 Causal Modeling ×1 Training Curriculum ×1 Sequential Social Dilemmas ×1 LLM-as-Policy ×1 Personalized Agents ×1 Code-Driven Generation ×1 Controllable Image Synthesis ×1 Layer-wise Editing ×1 Capability Analysis ×1 E4SRec ×1 KG-RAG ×1 Scalable Oversight ×1 Conformal Decision Theory ×1 Attainable Utility Preservation ×1 knowledge sources ×1 source-native query ×1 unified interface ×1 natural language query ×1 DMPO ×1 Zero-dependency Synthesis ×1 Modality Substitution ×1 Cross-modal Invariance ×1 Carrier Sensitivity ×1 data analysis ×1 Backdoor Attack ×1 Supply Chain Security ×1 Adapter Detection ×1 Fine-tuning Poisoning ×1 LLM Ranker ×1 K-means Clustering ×1 Embedding-based Ranking ×1 Report Generation ×1 Belief Entropy ×1 decision-focused learning ×1 PV forecasting ×1 battery scheduling ×1 Video Geometry Estimation ×1 Dynamic Chunking Attention ×1 Depth Completion ×1 CBM ×1 Belief Tracking ×1 Over-soliciting ×1 Orthogonal Finetuning ×1 Representation Geometry ×1 phone-use agent ×1 environment construction ×1 mock app ×1 4D Generation ×1 Human-Object Interaction ×1 Physical Simulation ×1 Motion Diffusion ×1 Affective Computing ×1 Wikipedia Co-occurrence ×1 Lightweight Verifier ×1 Checkpoint Repair ×1 Verified Replay ×1 Suffix Repair ×1 task assignment ×1 capacity constraints ×1 Agentic Data Engineering ×1 Model Specialization ×1 Explicit Feedback ×1 Source-Dependence ×1 Text-Guided ×1 Conditional Generation ×1 Ad Prediction ×1 rare-task learning ×1 memory retention ×1 OLMo ×1 Shortcut Bias ×1 Action-World Interaction ×1 embedding space ×1 mechanistic interpretation ×1 sparse features ×1 chain-of-thought internalization ×1 Structured Decoding ×1 Trigger Token ×1 Constrained Generation ×1 In-Writing ×1 Distributional Perturbation ×1 Vulnerability ×1 WASH ×1 Interpretable Reasoning ×1 VisAnomBench ×1 multimodal contrastive learning ×1 utility-aware ×1 product image generation ×1 demand model ×1 inverted U-shape ×1 Vision-LLM ×1 Mobile Automation ×1 Behavior Exploration ×1 Tone Sensitivity ×1 Model Robustness ×1 Embedding Anisotropy ×1 LLM-as-Extractor ×1 Linear Autoencoder ×1 Spectral Shrinkage ×1 misalignment ×1 developer-agent interaction ×1 real-world sessions ×1 IDE vs CLI ×1 AI Skills ×1 Trace-to-Skill ×1 Trojan Attack ×1 Multi-step Attack ×1 MaskNet ×1 embedding scaling ×1 search ranking ×1 multimodal document retrieval ×1 layout-aware sparse embedding ×1 few-shot learning ×1 GCPO ×1 Classifier-Free Guidance ×1 Multimodal RL ×1 compositional incoherence ×1 multi-component agents ×1 coherence monitoring ×1 projection repair ×1 e-process ×1 Shell Command Retrieval ×1 Parallel Execution ×1 hierarchical retrieval ×1 document QA ×1 coarse-to-fine ×1 evidence subgraph ×1 Model-Aware Skill Alignment ×1 Hill Climbing ×1 UCB Tree Search ×1 Skill Rewriter ×1 Power Distribution ×1 Entropy-Guided ×1 Decision Points ×1 Program Distillation ×1 Data Organization ×1 Folding ×1 Sharpening ×1 working memory ×1 memory blocks ×1 Source Bias ×1 Logical Fallacies ×1 LLM Robustness ×1 Distractor Construction ×1 Document-Level Translation ×1 Multi-Granularity Memory ×1 Embedding Transfer ×1 Power Law ×1 MemFT ×1 Phase Transition ×1 Curve Parameterization ×1 Semi-structured KB ×1 Dense Retriever ×1 Plan-guided ×1 Memory Policy ×1 Streaming VQA ×1 Adapter Tuning ×1 Token Probabilities ×1 skill bank curation ×1 Self-supervised RL ×1 Agent ToM ×1 dense retrievers ×1 latent vocabulary ×1 neural IR ×1 Online Serving ×1 Co-Evolving Policies ×1 Rubric-based Judging ×1 Detector Allocation ×1 Safety–Utility Trade-off ×1 Pre-hoc Reasoning ×1 preference adaptation ×1 human response time ×1 test-time finetuning ×1 convex approximation ×1 Frank-Wolfe ×1 gradient caching ×1 Prefix Certification ×1 Matryoshka Representation ×1 Preplan ×1 Plan-based Reasoning ×1 Self-anchored drift ×1 Canonical-context consistency ×1 Multi-turn dialogue ×1 looped transformer ×1 dual-path block ×1 per-token routing ×1 compute-capacity scaling ×1 concept vector ×1 Hysteretic Learning ×1 Temporal Graph Learning ×1 Trigger Detection ×1 On-device Deployment ×1 verifier-in-the-loop training ×1 Atomic Decomposition ×1 Evolving ×1 semantic factor learning ×1 alignment and uniformity ×1 false negative mitigation ×1 Prefix Encoder ×1 Hash Memory ×1 Dynamic Graph ×1 Collaborative Learning ×1 Retrieval Tree ×1 News Retrieval ×1 Subtree Selection ×1 Time-sensitive Retrieval ×1 Semantic Expansion ×1 modular addition ×1 Text-based Recommendation ×1 Embedding Tying ×1 Multi-Objective Bandits ×1 Public Media ×1 Graph-GRPO ×1 Generative Relevance ×1 Numerical Hallucination ×1 Data-centric Compilation ×1 Structuring Agent ×1 Answer Retention ×1 Multi-turn optimization ×1 Weighted SFT ×1 Importance sampling ×1 Masked Diffusion Language Models ×1 Graph-to-Text ×1 BLEU ×1 Search Tree ×1 Parent Pointers ×1 Structured Trace ×1 zero-shot annotation ×1 prior ×1 correction stickiness ×1 DSF ×1 toxicity detection ×1 visual document retrieval ×1 caption-gated supervision ×1 lexical grounding ×1 inference-free ×1 Conditioning ×1 Balanced ×1 Agent-native ×1 ATE-Bench ×1 training framework ×1 LOOP Module ×1 Misleading Queries ×1 Novelty Detection ×1 Write-side Control ×1 Rewarded Perturbations ×1 Low-Rank Structure ×1 Gradient-Free ×1 scalable recommendation ×1 Unified Retrieval and Ranking ×1 Masked Action Modeling ×1 KV Cache Sharing ×1 Production Recommendation ×1 Visual Skills ×1 Textual Bottleneck ×1 Skill Adaptation ×1 Step-level Attribution ×1 Failure Localization ×1 trust functions ×1 snowballing chain ×1 Small Reasoning Models ×1 predictive coding ×1 video MLLM ×1 visual token compression ×1 adaptive GOP ×1 motion-residual encoding ×1 security scanning ×1 scanner disagreement ×1 skill governance ×1 Anonymization ×1 Agentic Re-identification ×1 Mask-Reconstruct ×1 multi-domain RL ×1 parameter sparsity ×1 second-order damage ×1 refresh training ×1 task-decomposition ×1 parallel-execution ×1 test-time-scaling ×1 geometric embedding ×1 length reduction ×1 emergent phenomenon ×1 Layer Skipping ×1 Straight-Through Estimator ×1 Ads CTR ×1 RQ-FSQ ×1 Skill Poisoning ×1 Attack Taxonomy ×1 Automated Construction ×1 Code Coverage ×1 Text-Visual Interleaved Report ×1 Visual Evidence Grounding ×1 Temporal Knowledge Graph ×1 Invalidation ×1 Interactive-Native ×1 Economic Coordination ×1 Decentralized Intelligence ×1 semi-supervised learning ×1 image segmentation ×1 medical imaging ×1 pseudo-label reweighting ×1 World Modeling ×1 On-policy RL ×1 MAE Loss ×1 Setwise ×1 DualEnd ×1 Submodule Replacement ×1 Residual Bypass ×1 semantic personas ×1 late-fusion ×1 multi-agent reinforcement learning ×1 data science agent ×1 EvoTrainer ×1 Training Harness Co-evolution ×1 Diagnostic Gap ×1 Stateful Harness ×1 Cognitive Offloading ×1 harness-policy co-evolution ×1 meta-adaptation ×1 Pareto selection ×1 CAPTCHA ×1 Preference Disentanglement ×1 Time-Aware Masking ×1 Agent Workflow ×1 Trajectory Verification ×1 Dependent Types ×1 Auto-Evolution ×1 Overlap-Aware Retrieval ×1 Evidence Quality ×1 ODTQA ×1 LLM-based Framework ×1 Persuasion ×1 RL Profile Analyzer ×1 Pre-fusion Calibration ×1 Cross-modal Comparison ×1 Modality Reliability ×1 Plug-in Module ×1 PTZ camera ×1 edge deployment ×1 agentic control ×1 image compression ×1 cloud robotics ×1 JPEG compatibility ×1 learned quantization ×1 relevance prior ×1 document findability ×1 neural retrievers ×1 annotation bias ×1 Temporal-Logic VideoQA ×1 Evidence-Seeking Agent ×1 Multi-Granular Sampling ×1 Gemini 3.1 Pro ×1 Self-supervised Skill Mining ×1 Long-horizon Task ×1 Safety Constraints ×1 Dynamic Environments ×1 Skill Boundary ×1 Selective Reuse ×1 Graph-Augmented Retrieval ×1 Evidence Sufficiency Scoring ×1 Self-Correcting Generation ×1 Prompt Distillation ×1 Small Models ×1 Rust ×1 Affine Types ×1 Incident Catalog ×1 Scaffold ×1 shadow price ×1 constrained optimization ×1 Policy Diversity ×1 Small-to-Large ×1 Forecast Revision ×1 Memory Reflection ×1 Map-Reduce Decomposition ×1 Capability System ×1 Runtime ×1 OS Metaphor ×1 Black-box Distillation ×1 Model Extraction ×1 Agent-Native Interface ×1 CLI Harness ×1 GUI Replacement ×1 Stateful CLI ×1 graph extraction ×1 logic puzzles ×1 structural analysis ×1 Deontic Reasoning ×1 Agentic Harness ×1 Reasoning Benchmark ×1 Capability Amplifier ×1 dynamic convolution ×1 architecture ×1 triton kernel ×1 Communication Protocol ×1 Action-State ×1 PACT ×1 Multi-turn Tool Calling ×1 Programmatic Reward ×1 skill compatibility ×1 Clinical Agents ×1 Standardized Patients ×1 online skill learning ×1 state-grounded retrieval ×1 sliding-window extraction ×1 MMR reranking ×1 Multi-rate aggregation ×1 Temporal dynamics ×1 Sequential recommendation ×1 Data density adaptation ×1 stochastic ×1 value-aware ×1 Personal AI Agent ×1 Camera Roll VQA ×1 Long-context Understanding ×1 Spatial Prediction ×1 MeRa ×1 System Prompt Optimization ×1 Self-Referential ×1 entropy quenching ×1 fidelity gate ×1 agentic workflows ×1 energy-squared amplification ×1 Resource Orchestration ×1 Semantic-ID ×1 Hidden Problem Discovery ×1 Thought Templates ×1 Iterative Discovery ×1 Parameter Generation ×1 knowledge-intensive reasoning ×1 training corpus ×1 Cognitive Amortization ×1 Curvature ×1 Normalized Directional Sharpness ×1 bias injection ×1 environment ×1 Real-world tasks ×1 Generalist CUA ×1 Reasoning Steering ×1 Step-wise Injection ×1 Constraint Elicitation ×1 role-playing ×1 character-arc ×1 language-agents ×1 narrative-understanding ×1 Scientific Collaboration ×1 On-premise LLM ×1 Async RL ×1 Multi-step RL ×1 Visual Web Agent ×1 Off-policy Correction ×1 Implicit-Need Surfacing ×1 Proactive Probing ×1 Situated Agents ×1 closed-loop optimization ×1 persistence ×1 iteration ×1 Attribute Taxonomy ×1 Prompt Refinement ×1 Gradient Masking ×1 causal reasoning ×1 lexical anchors ×1 benchmark perturbation ×1 placeholder tokens ×1 Linear Bandits ×1 Offline-to-Online ×1 LCB ×1 Warm-start ×1 Code Language Model ×1 Execution Provenance ×1 Agent Observability ×1 3D Geometry ×1 Video Representation ×1 Code Documentation ×1 Cross-File Dependencies ×1 modular training ×1 Cascading Hallucination ×1 Multi-Step Reasoning ×1 Discrete Semantic Identifiers ×1 Contrastive RQ-VAE ×1 LLM for Query Understanding ×1 Position-aware Calibration ×1 Uncertainty-Aware Retrieval ×1 Long-tail Coverage ×1 Embedding Uncertainty ×1 Forward-Looking Judgment ×1 Complexity-Balanced Splitting ×1 Temporal Partitioning ×1 De Boor Equidistribution ×1 validity threats ×1 proxy experiments ×1 observational studies ×1 research methodology ×1 LLM Explainability ×1 Causal Graphs ×1 Counterfactual Augmentation ×1 Concept-based Explanations ×1 Forward Cross-Entropy ×1 Monotonic Improvement ×1 persuasion ×1 belief tracing ×1 human-LLM interaction ×1 Bayesian network ×1 simulator ×1 Continuous Thoughts ×1 Sales Lead Scoring ×1 low-resource translation ×1 chrF reward ×1 PPI ×1 Precision@K ×1 Ranking Evaluation ×1 Hierarchical Metrics ×1 Structured Defect Grounding ×1 T2I Diagnosis ×1 SDG-30K ×1 Recall@k ×1 1-Ratio@k ×1 function vectors ×1 LRP ×1 Reflective Agent ×1 Edit Propagation ×1 dynamic-inference ×1 layer-skipping ×1 layer-looping ×1 adaptive-computation ×1 Propensity ×1 Efficient Architecture ×1 MCGS ×1 representation alignment ×1 query operators ×1 typed events ×1 IE-to-SQL ×1 Benchmark Agent ×1 Autonomous Benchmarking ×1 Disentangled Representation ×1 Pre-extracted Embeddings ×1 Attention Steering ×1 Trajectory Features ×1 Post-Training Analysis ×1 vision-language-action ×1 autoregressive transformer ×1 robot control ×1 Clinical Summarization ×1 Expert Evaluation ×1 Blinded Study ×1 Dual-Stream MLP ×1 Alignment Loss ×1 Protocol ×1 Langshaw ×1 UCP ×1 Declarative ×1 Streaming Communication ×1 Step-Level Scaling Law ×1 carbon-aware recommendation ×1 product carbon footprint ×1 LLM for PCF estimation ×1 Pareto trade-off ×1 Prefix Tuning ×1 Temporal Attention ×1 MADDPG ×1 Action Inference ×1 Non-Stationarity ×1 TDA ×1 sparse recovery ×1 parametric retrieval ×1 diagnostic benchmark ×1 knowledge retention ×1 internalization score ×1 Skill Discovery ×1 Video Event Prediction ×1 Behavioral Foundation Models ×1 Event Embedder ×1 Compute-Optimal Allocation ×1 Early Failure Alerting ×1 Weak Supervision ×1 Multi-turn Dialog ×1 Preference-Conditioned RL ×1 Repeated Games ×1 Adaptive Opponents ×1 Multi-Agent Learning ×1 Non-convex Optimization ×1 API design ×1 error recovery ×1 structured feedback ×1 self-reflective API ×1 self-evaluation ×1 elicitation ×1 Empathy ×1 Emotional Support ×1 Verbal Harassment ×1 Coping ×1 Recommendation Denoising ×1 Agent-based Simulation ×1 LLM-as-User ×1 Supervised Denoising ×1 Creator-Recognizer ×1 in-band signaling ×1 access control ×1 recuse signal ×1 agent compliance ×1 Collaborative Competence ×1 CSCW ×1 Hate Speech ×1 Perspective-Taking ×1 Demographic Alignment ×1 Vicarious Prompting ×1 Annotation Disagreement ×1 Asynchronous Training ×1 Weight Staleness ×1 Gradient Accumulation ×1 Risk-Averse Decision ×1 Action-Conditional Guarantee ×1 Distribution-Free ×1 Nash Equilibrium ×1 Bidding ×1 Multimodal retrieval ×1 Active modality detection ×1 Query-adaptive fusion ×1 Speaker embedding ×1 Face embedding ×1 Broadcast archive ×1 factual sycophancy ×1 truth margin ×1 manipulation sensitivity ×1 Multi-behavior Recommendation ×1 KAN ×1 Self-preference ×1 memory reconstruction ×1 graph memory ×1 active retrieval ×1 World Simulator ×1 Multimodal Sentiment Analysis ×1 Discriminative Readout ×1 Generative Readout ×1 Qwen2.5-Omni ×1 Hidden-State Regression ×1 cross-layer sharing ×1 Self-Augmenting ×1 ℓ2 norm ×1 reasoning dynamics ×1 deployment strategy ×1 Video Codec ×1 Affine Transform ×1 VVC ×1 Multi-Vertical Recommendation ×1 MTL ×1 Asymmetric Architecture ×1 Link Prediction ×1 Content Embedding ×1 lexical density ×1 needle-in-a-haystack ×1 information density ×1 Graph LLM ×1 Edge Curvature ×1 Over-squashing ×1 Node Alignment ×1 Curvature-aware Prompt ×1 Molecular Property Prediction ×1 Chemistry ×1 Structure Retrieval ×1 Itemic Token ×1 Perception-Cognition ×1 Multi-Domain Rec ×1 Preconditioning ×1 Weight Normalization ×1 Singular Value Spectrum ×1 mental model ×1 human-agent collaboration ×1 action annotation ×1 Temporal Redundancy ×1 Reward Redistribution ×1 Perceiver ×1 Double Preconditioning ×1 Test-Time Feedback ×1 Activation Preconditioning ×1 Efficiency Optimization ×1 Step-Level MDP ×1 Belief Revision ×1 Structural Graph ×1 TailLoR ×1 Spectral Decomposition ×1 Interference ×1 Video Moderation ×1 Interpretable AI ×1 Content Safety ×1 programmable system ×1 paged tensor ×1 SEO ×1 Internal Linking ×1 GraphSAGE ×1 Semantic Coherence ×1 Link Evaluation ×1 SMT ×1 RNN ×1 long-range dependencies ×1 gradient stability ×1 misconception diagnosis ×1 knowledge-gap localization ×1 interpretable AI ×1 LLM Grading ×1 Rubric Faithfulness ×1 Intervention Training ×1 Belief-Shaped RL ×1 Internal-State Diagnostics ×1 Adaptive Orchestration ×1 Systems Characterization ×1 Knowledge Infusion ×1 Layered Intervention ×1 Multimodal Knowledge Graph ×1 LLM personalization ×1 synthetic vs real ×1 user attribute extraction ×1 relevance matching ×1 personalized response ×1 Life-Long Learning ×1 Social Intelligence ×1 knowledge work ×1 task scope ×1 natural experiment ×1 Bayesian Agent ×1 Posterior-Guided ×1 cheating detection ×1 capped evaluation ×1 Byzantine resilience ×1 semantic commitment ×1 typed finality ×1 multi-agent collaboration ×1 Prefix Evaluation ×1 Gain-based Utility ×1 Cross-modal Coordination ×1 Fisher-Rao Geodesic ×1 Multi-Path Inference ×1 Graph-Based Planning ×1 Rubric-Guided Reasoning ×1 Recursive Execution ×1 Closed-form Solver ×1 Data-free ×1 multimodal QA ×1 min-cost flow ×1 evidence assembly ×1 long document ×1 Category Hierarchy ×1 SWE-agent ×1 trace distillation ×1 skill registry ×1 Multi-hop RAG ×1 Key-value retrieval ×1 Evidence indexing ×1 ColBERTv2 ×1 Unembedding Matrix ×1 Spectral Filtering ×1 Dimensionality Reduction ×1 Reversible Architecture ×1 State-Preserving Growth ×1 Quantized Training ×1 Single-Node Training ×1 Evolutionary Algorithms ×1 Medical Decision Pipelines ×1 LLM-guided Evolution ×1 Modality-aware Retrieval ×1 interaction space ×1 retrieval interface ×1 sparse experts ×1 Activation Outliers ×1 Offsetting ×1 Top-1 PCA ×1 Paper Recommendation ×1 Interest Drift ×1 Multi-signal Ranking ×1 LLM Cascading ×1 Pandora’s Box ×1 API Selection ×1 visual self-recovery ×1 image corruption ×1 3D Spatial Reasoning ×1 MLLM Agents ×1 Self-Evolving Skills ×1 Scene-Aware ×1 Hybrid Encoder ×1 Long Sequence ×1 Live Streaming Recommendation ×1 Dynamic Semantic ID ×1 Generative-Discriminative Architecture ×1 Pre-Ranking ×1 IQL ×1 Residual Connection ×1 Dual-Path Network ×1 Prefix failure ×1 Trajectory refinement ×1 Video RAG ×1 Persona Adherence ×1 Writing Assistant ×1 Overleaf ×1 Masking ×1 Token Subset ×1 Closed-Loop Self-Enhancement ×1 context-compression ×1 soft-tokens ×1 LLM-inference ×1 Multi-Condition Retrieval ×1 Semantic Dilution ×1 Stage-Aware Framework ×1 Hard Negative Discrimination ×1 Neural Memory Indexer ×1 Item-Level Scoring ×1 Collision Resolution ×1 Particle Filter ×1 Verification-Free ×1 Intrinsic Selection ×1 adversarial hardening ×1 agent benchmarks ×1 verifier hardening ×1 weak-to-strong defense ×1 VectorSearch ×1 Block-SoA ×1 Tangent-Local Projection ×1 Reward Noise ×1 Game Benchmark ×1 UE5 ×1 Sign Consistency ×1 Verifier-guided ×1 Role-Decoupled Optimization ×1 Single-Policy ×1 On-Device Assistant ×1 KTO ×1 GUI-CLI Hybrid ×1 Long-horizon Tasks ×1 Trajectory-aware Evaluation ×1 Arbitration ×1 divergence regularization ×1 DRPO ×1 Prefix Constraint ×1 Position-Weighted ×1 Token Masking ×1 omni-modal retrieval ×1 embedding model ×1 model fusion ×1 projector drift ×1 feedback alignment ×1 step-aligned critique ×1 highlighting ×1 salience ×1 selection ×1 co-readership ×1 QK-Restore ×1 CoT-SFT ×1 hybrid linear attention ×1 long-context recall ×1 gradient locality ×1 Feature Alignment ×1 mobile agents ×1 multi-app reasoning ×1 Long-Form Generation ×1 Dynamic Planning ×1 behavior forecasting ×1 reasoning trajectory ×1 Structured Video Understanding ×1 Logic-aware Debate ×1 Counterfactual Explanations ×1 Glucose Management ×1 Personalized Nutrition ×1 Wearable Sensing ×1 Expert-in-the-loop ×1 pacing ×1 advertising ×1 decision calibration ×1 dual-weighted score ×1 Set-wise preferences ×1 Multinomial surrogate ×1 Embedding Mixing ×1 recommendation benchmark ×1 visual evidence ×1 movie recommendation ×1 Decentralized Coordination ×1 Delegation Intelligence ×1 pass^k ×1 Visual Feature Learning ×1 Modality Imbalance ×1 Gradient Reweighting ×1 Prefix Retention ×1 PageMem ×1 Compound Actions ×1 failure audit ×1 decoy calibration ×1 error slice discovery ×1 model debugging ×1 lift ×1 holdout replication ×1 feature stability ×1 seed dependence ×1 low-rank subspaces ×1 epistemic resilience ×1 misleading context ×1 medical LLM evaluation ×1 Multimodal Image Fusion ×1 Token Representation ×1 Selective Token Editing ×1 Shared Representation ×1 web measurement ×1 incrementality ×1 consumer journey ×1 TV Loss ×1 Agentic Environment Engineering ×1 Prompt Learning ×1 Router-Prompt Co-evolution ×1 Tool Agent ×1 Workflow Evolution ×1 Inference-Time Search ×1 Compact LM ×1 vector search dilution ×1 domain scoping ×1 Zero-Variance ×1 Query Recycling ×1 MCR ×1 M2LA ×1 Social World Model ×1 belief dynamics ×1 prediction markets ×1 state transition ×1 Router ×1 Power Iteration ×1 Environment Composition ×1 Reasoning Generalization ×1 Verifiable Environments ×1 Recursive Composition ×1 activation_steering ×1 steerability_prediction ×1 inference_time_control ×1 early_decoding ×1 Latent Space Retrieval ×1 Visual Cache ×1 Parallel Tempering ×1 Omnimodal ×1 Agent Swarm ×1 Compatible Learning ×1 Stationary Representation ×1 d-Simplex Classifier ×1 Contrastive Loss ×1 Dual-Role ×1 Failure Mode Retrieval ×1 item representation ×1 archetype ×1 behavioral calibration ×1 VLM reward ×1 Event-Driven ×1 Delegate ×1 AdaCodec ×1 Data Journalism ×1 Agent-as-Judge ×1 tokenizer artifacts ×1 diagnostic evaluation ×1 mapping inspection ×1 prefix alignment ×1 same-capability ambiguity ×1 utility learning ×1 representative selection ×1 Lexical Expansion ×1 Reward Guidance ×1 multi-domain ×1 Hypothesis Tree Refinement ×1 Adversarial Learning ×1 Experiential Knowledge ×1 Knowledge-Augmented ×1 Multi-turn RLVR ×1 Tree-structured Rollouts ×1 Prefix Value Prediction ×1 Trial-and-Error ×1 Hybrid Simulation ×1 Metrics ×1 LLM Dependency ×1 Model Supply Chain ×1 Documentation ×1 Credit_Assignment ×1 Information_Flow ×1 Cross-Domain Recommendation ×1 Intent Modeling ×1 Self-Gated ×1 Self-Report ×1 TPB ×1 Big 5 ×1 Counterfactual Error Control ×1 Tool Grounding ×1 CopyNet ×1 Attention Scoring ×1 skill compression ×1 procedural knowledge ×1 multi-resolution ×1 Daily Tasks ×1 Data Debugging ×1 Multi-Label Classification ×1 Set Transformer ×1 parametric memory ×1 atom decomposition ×1 long-document reasoning ×1 Budget-aware ×1 Dynamic Environment ×1 Patch-based Memory ×1 evolving knowledge ×1 multi-agent synthesis ×1 contamination-free ×1 Multimodal Entity Linking ×1 Vector Representation ×1 Shortcut-Resistant ×1 Trajectory Signatures ×1 Runtime Enforcement ×1 User Correction ×1 Access-Compliance Gap ×1 HarnessBridge ×1 Observation Projection ×1 Action Projection ×1 Visual Tokenizer ×1 ViT ×1 Latent Space ×1 Planner-Critic ×1 APPO ×1 Branching ×1 Procedural Reasoning ×1 User Persona ×1 Multi-turn RAG ×1 Unanswerable Queries ×1 Hidden-State Recurrence ×1 Switchable Reasoning ×1 Mechanistic Analysis ×1 RL for LLM ×1 Scaling Factor ×1 Heterogeneous Agents ×1 Dense Alignment ×1 Retrieval-enhanced Aggregator ×1 Seesaw Dilemma ×1 Top-k Retrieval ×1 Embedding Dimension ×1 Video Copy Detection ×1 Recommendation Pipeline ×1 Generative Verifier ×1 Proof Generation ×1 code retrieval ×1 embedding models ×1 Grouped Query Attention ×1 Top-k Selection ×1 video deduplication ×1 multi-level embeddings ×1 spatial-temporal matching ×1 near-duplicate detection ×1 large-scale ×1 Decentralized Learning ×1 Online Learning to Rank ×1 Byzantine Robustness ×1 Gossip Learning ×1 Code as Action ×1 Iterative Reasoning ×1 Hint Optimization ×1 Intra-Model Routing ×1 Hierarchical Verification ×1 Extreme Value Theory ×1 Knee Detection ×1 Language Steering ×1 phase transition ×1 Bayesian theory ×1 induction head ×1 Visual Question Generation ×1 reward fine-tuning ×1 any-length generation ×1 adaptive decoding ×1 retrieval-grounded feedback ×1 iterative refinement ×1 visual autoregressive ×1 bitwise-residual ×1 Bernoulli guidance ×1 residual code ×1 operadic consistency ×1 compositional reasoning ×1 Fashion Recommendation ×1 Outfit Generation ×1 Hybrid Encoding ×1 Commitment Boundary ×1 Epiphenomenal Reasoning ×1 Attention Probe ×1 constraint compatibility ×1 negative constraints ×1 lexical-polarity supervision ×1 antonymy-negation-exclusion ×1 Large Language Model ×1 Proactive Investigation ×1 Task Closure ×1 Cross-Query Consistency ×1 Recursive Agent Harness ×1 Workspace Expansion ×1 Turn-taking ×1 Multi-party ×1 multilingual dense retrieval ×1 query embedding interpolation ×1 code-mixing ×1 BGE-M3 ×1 typological distance ×1 Multiagent Debate ×1 Confidence Aggregation ×1 Bayesian Fusion ×1 LLM Collaboration ×1 Repository Exploration ×1 Parallel Tool Calls ×1 bilevel optimization ×1 gradient descent ×1 Dynamic Parallelism ×1 GPU Scheduling ×1 vLLM-Omni ×1 Group-Free Collective ×1 AEGIS ×1 Dialogue Evaluation ×1 Dataset Generation ×1 Tool-Augmented Agents ×1 Execution Abstraction ×1 Data Attribution ×1 CAD Generation ×1 Geometric Reasoning ×1 counterfactual learning ×1 network management ×1 Dispatch ×1 Objective-Weight Adaptation ×1 Marketplace ×1 User-space I-O ×1 Learned Pruning ×1 positional encodings ×1 truncation ×1 expressive power ×1 WL test ×1 Code Repository ×1 Vision ×1 Operads ×1 sparsity ×1 parameter analysis ×1 Document Chunking ×1 Entropy-based Fusion ×1 adaptive token compression ×1 frequency domain ×1 asymmetric token ×1 common sense ×1 Hybrid Mamba-Attention ×1 Keyword-Aligned Encoding ×1 Multi-Branch Retrieval ×1 Editability ×1 Phone Agent ×1 Mixed Actions ×1 Verifiable Execution ×1 Granularity ×1 Hierarchical ×1 Certified Robustness ×1 Hessian Regularization ×1 Adversarial Defense ×1 stakeholder-centric ×1 ensemble pruning ×1 bagging ×1 probability calibration ×1 simplex constraint ×1 concave regularization ×1 iterative retrieval ×1 social highlighting ×1 temporal stability ×1 reading identity ×1 Analogy Reasoning ×1 Self-Verified Training ×1 Representational Geometry ×1 Subspace Analysis ×1 Accessibility Collapse ×1 Web-Content Pollution ×1 Generative Recommenders ×1 LLM Vulnerability ×1 Search-Augmented Generation ×1 Fake Reviews ×1 FORGE Benchmark ×1 model editing ×1 machine unlearning ×1 Granger Causality ×1 Cloud Networks ×1 Graphical Models ×1 Fault Diagnosis ×1 Attribute-aware Learning ×1 Zero-shot Retrieval ×1 Parallel Quantization ×1 Relative Time ×1 semi-online RL ×1 reward alignment ×1 silent failures ×1 fail-plausible ×1 governance ×1 postmortem analysis ×1 Denoising Recommendation ×1 Small-loss Heuristic ×1 PAD ×1 Progress Awareness ×1 Retrospective Refinement ×1 Decoupled MoE ×1 Uncertainty-aware Routing ×1 Parametric Knowledge ×1 Architecture Co-design ×1 Preference Coordination ×1 CTDE ×1 Synthesis ×1 Parallel Workflow ×1 heavy-tailed noise ×1 non-Euclidean optimization ×1 Transformer training ×1 Gaze Steering ×1 Comic Strips ×1 Score Fusion ×1 Chunk Selection ×1 Latent Failures ×1 memory evaluation ×1 streaming benchmark ×1 evidence reuse ×1 future-oriented assistance ×1 lifelog ×1 file-based protocol ×1 heterogeneous agents ×1 self-applied operation ×1 deep search ×1 active visual reasoning ×1 action sequence abstraction ×1 hierarchical prompting ×1 intent understanding ×1 memory-augmented ×1 R-GCN ×1 multi-modal memory ×1 efficient attention ×1 retrieval heads ×1 Implicit Negative Feedback ×1 Target-Aware Polarity Fusion ×1 Behavior Sequence ×1 Attention Rebalancing ×1 Hierarchical Semantic Encoding ×1 Interleaved Reasoning ×1 Generative Query Recommendation ×1 PIOPD ×1 Preference Internalization ×1 TrajectoryConstruction ×1 Video Object-Centric Learning ×1 Selective Alignment ×1 Pseudo-labeling ×1 Object Discovery ×1 economic simulation ×1 long-horizon decision making ×1 supply chain ×1 E-PRoPE ×1 Evidence Budget ×1 Mask Diffusion Models ×1 Reflective Masking ×1 Local Edits ×1 History Reference ×1 Helpfulness-Safety ×1 PFN ×1 Synthetic Prior ×1 Support Set ×1 Update-Free ×1 over-specialization ×1 biology ×1 Ad Description Generation ×1 Iterative Creation ×1 Knowledge Capacity ×1 Landing Page Consistency ×1 Implicit Reasoning ×1 Pause Tokens ×1 Multi-turn LLM ×1 Non-uniform Compression ×1 Serving Efficiency ×1 gradient bank ×1 weight drift ×1 no-resource language ×1 weight diff transfer ×1 Multi-modal Generation ×1 Feature Fusion ×1 Perceptual Decoding ×1 Lightweight Decoder ×1 Desktop Automation ×1 Web Tasks ×1 Gradient Detachment ×1 Candidate-Aware ×1 Dynamic Scoring ×1 Disentangled Representations ×1 Multi-modal Fusion ×1 Edit Consistency ×1 WAPO ×1 VLM Reasoning ×1 Synthetic Data Pipeline ×1 Point-Box Grounding ×1 Gemma ×1 Decoding Dynamics ×1 Knowledge Fusion ×1 Iterative Decoding ×1 initialization ×1 Taylor expansion ×1 Lifecycle-Aware ×1 Multimodal Instruction ×1 Drag Editing ×1 Dataset Construction ×1 Startup Simulation ×1 Strategic Planning ×1 Partially Observable ×1 parallel sampling ×1 query diversity ×1 Multimodal Code Intelligence ×1 Visual-to-Code ×1 GUI Generation ×1 Agentic Code ×1 game generation ×1 Godot ×1 interactive verification ×1 multimodal judging ×1 trajectory-level supervision ×1 preference feedback ×1 procedural representation ×1 ProcGrep ×1 Width ×1 FLOPs ×1 hallucination auditing ×1 legal AI ×1 risk direction index ×1 typed claims ×1 adaptive sampling ×1 Retriever-Aware ×1 LLM Memorization ×1 One-hop Memorization ×1 Item-Item Relations ×1 Auxiliary Training ×1 LLM search agents ×1 endorsement vulnerability ×1 web content manipulation ×1 Non-Monotonic Scaling ×1 Gain-Cost Trade-off ×1 agentic streaming ×1 reinforcement distillation ×1 Slide Generation ×1 Local Revision ×1 Tool Memory ×1 Expert Tying ×1 Experience Lifecycle ×1 Skill Tree ×1 Multi-Model ×1 personalized reranking ×1 listwise reranking ×1 information aggregation node ×1 music search ×1 Spatial VLM ×1 Dual-Path Reasoning ×1 time-series forecasting ×1 Collaborative Reward ×1 RL-based Collaborative Injection ×1 index reasoning ×1 retrieval augmentation ×1 Semantic Continuity ×1 Context Expansion ×1 Adaptive Chunking ×1 Context-Aware RL ×1 Contrastive Context ×1 Agentic ×1 Code Interpreter ×1 Crucial Tokens ×1 Cognitive Behaviors ×1 Discrete Visual Tokenizer ×1 Diffusion Decoder ×1 Looped Architecture ×1 ZPPO ×1 Small Model Training ×1 reinforcement-learning ×1 latent modules ×1 training strategy ×1 Trajectory Construction ×1 rollout editing ×1 pipeline optimization ×1 autonomous agent ×1 Claude Code ×1 multi-step LLM ×1 structural bottleneck ×1 Access Control ×1 Active Forgetting ×1 Conformal Risk Control ×1 Drift Detection ×1 Anytime Alarm ×1 L0 Regularization ×1 Lagrangian Optimization ×1 3D control ×1 DNOCS ×1 spatial blocking ×1 layout conditioning ×1 Topic Metadata ×1 Paragraph-level ×1 system-aware ×1 rollout acceleration ×1 Omni-Modal Agent ×1 Unsupervised Dense Retriever ×1 TRPO ×1 Time Embedding ×1 Scientific Reasoning ×1 Variational Free Energy ×1 Energy-based Model ×1 Lightweight Architecture ×1 Image Inpainting ×1 SEAL ×1 MINDER ×1 debugging tool ×1 OpenRath ×1 Runtime Abstraction ×1 graph learning ×1 co-design ×1 billion-node ×1 popularity bias correction ×1 cluster index ×1 U2U2I retrieval ×1 sequential revision ×1 multi-turn RL ×1 off-policy data ×1 replay buffer ×1 Representational Shift ×1 Relevance Scoring ×1 Lightweight Training ×1 信息茧房 ×1 闭环仿真 ×1 LLM用户模拟 ×1 代码空间塌缩 ×1 Attention-Guided Chunking ×1 UDLM ×1 7B ×1 sign patterns ×1 feature discovery ×1 dimension-level ×1 shortcut mitigation ×1 perception-reasoning decoupling ×1 Discriminator-Guided RL ×1 FID Reduction ×1 KL-Regularized RL ×1 Preference-Free Reward ×1 Bi-encoder Retrieval ×1 Index Sickness ×1 Semantic Vitality ×1 AI Instructions Reduction ×1 Phantom Legislation ×1 Long-Horizon Collaboration ×1 Predictive Validity ×1 OOD ×1 Leaderboard ×1 Subspace Decomposition ×1 Low-rank Experts ×1 GQA ×1 sparse-attention ×1 kv-cache ×1 conditional-computation ×1 GraphPO ×1 semantic equivalence ×1 advantage estimation ×1 Trajectory Summarization ×1 Gibbs sampling ×1 structured inference ×1 LLM conditionals ×1 probabilistic reasoning ×1 Active Feature Acquisition ×1 Markov Random Field ×1 Maximum Entropy ×1 Cost Optimization ×1 Policy Adherence ×1 Tool-Calling Agents ×1 Multi-Turn Consistency ×1 AutoRecSys ×1 LensKit ×1 Tree Parzen Estimator ×1 Usability ×1 Agent Communication Protocols ×1 Schema Flexibility ×1 Protocol Stack ×1 E-commerce Video ×1 pseudo-labeling ×1 embedded vector search ×1 training-free quantization ×1 Randomized Hadamard Transform ×1 edge AI ×1 determinism ×1 Contamination ×1 Distribution Adaptation ×1 Tool-Use Agent ×1 Evidence Accumulation ×1 SFT Distillation ×1 Intent-based Recommendation ×1 Textual Data ×1 Stochastic Search ×1 Abstractive Summarization ×1 Reflective Refinement ×1 User Experience ×1 UX Principles ×1 Mixed-Methods ×1 Model-Based RL ×1 Surprisal ×1 Entropy Regulation ×1 Soft Token ×1 Feature Compression ×1 Large Recommendation Models ×1 PLUM ×1 YouTube ×1 Multi-Agent Fictitious Play ×1 Equilibrium-Seeking ×1 Stance Entanglement ×1 Decision-Making ×1 Steerable ×1 Positive-Unlabeled Learning ×1 cognitive diversity ×1 programming assistants ×1 interaction modes ×1 developer needs ×1 GitHub Copilot ×1 Decision-focused learning ×1 Predict-then-Optimize ×1 Linear Programming ×1 Surrogate Loss ×1 Solver-Free ×1 Selective Verification ×1 Budget-Aware Reasoning ×1 Verification Allocation ×1 Memory Edit ×1 Token-level Guidance ×1 Shrinkage Bias ×1 Uniform Grid ×1 Random Hadamard Transform ×1 MLM-head rescaling ×1 Marginal Advantage Accumulation ×1 Trace Distillation ×1 Offline Memory Optimization ×1 EMA Evidence Accumulation ×1 Semantic Identity Merging ×1 probe-and-refine ×1 repository guidance ×1 LLM tuning ×1 knowledge base ×1 Self-Confidence ×1 Grounding Verifier ×1 Sequential DPO ×1 Multi-objective ×1 Hierarchical Recovery ×1 Cross-Device Agent ×1 Oversensitivity ×1 Monte Carlo Dropout ×1 Discretization ×1 contagion-networks ×1 evaluator-bias ×1 multi-agent-llm ×1 propagation-regimes ×1 diversity-suppression ×1 spectral-radius ×1 process critique ×1 open-source dataset ×1 AI_Search_Visibility ×1 Brand_Visibility ×1 Content_Strategy ×1 Citation_Analysis ×1 hubness ×1 global gating ×1 anisotropy ×1 Agentic Data Tailoring ×1 Multimodal Refinement ×1 Factual Anchors ×1 Amortized Inference ×1 Transactive Memory ×1 Trajectory Retrieval ×1 RAG for Agents ×1 Grain Blindness ×1 Ranking Reward ×1 Evolvable Embeddings ×1 Adversarial Reweighting ×1 Soft Routing ×1 Hard Routing ×1 Fisher Information ×1 Riemannian Geometry ×1 data bias ×1 coverage constraints ×1 integer linear programming ×1 data efficiency ×1 intersectionality ×1 Graph Clustering ×1 Co-engagement Graph ×1 Soft Modularity ×1 filtered ANN ×1 query-aware routing ×1 ML-based router ×1 recall-QPS trade-off ×1 vector database ×1 Head-level Fusion ×1 Skill Mining ×1 Interaction Trajectory ×1 Diagnostic Study ×1 Fitted Q-Evaluation ×1 Bridge Function ×1 data-constrained ×1 Text Diffusion ×1 DiffusionGemma ×1 Timestep Embeddings ×1 long-lifecycle agents ×1 meta-RL ×1 end-to-end RL ×1 Tool-Intent Stabilization ×1 Latency Hiding ×1 Speculative Query ×1 CRAG Benchmark ×1 response bias ×1 psychometrics ×1 personality assessment ×1 measurement artifact ×1 Makeup Transfer ×1 Facial Attribute Preservation ×1 ControlNet ×1 Proactive ×1 Mobile ×1 PR-AUC ×1 precision-recall ×1 Probabilistic Policies ×1 Distributionally Robust Optimization ×1 Datalog ×1 Token Operations ×1 Model Quantization ×1 Hierarchical Feedback ×1 Curriculum Mining ×1 eye tracking ×1 mouse trajectory ×1 LLM agent safety ×1 multi-turn red-teaming ×1 jailbreak benchmarks ×1 safety-critical systems ×1 context-management ×1 submodular-optimization ×1 pluggable-engine ×1 facility-location ×1 Space Supervision ×1 Mutual Information ×1 Over-Privileged ×1 Post-Training Defense ×1 ToolPrivBench ×1 academic search ×1 Soft prompt tuning ×1 Prompt compression ×1 Agent skill ×1 Behavioral compression ×1 Prefix tuning ×1 Spatially Speculative Decoding ×1 Autoregressive Image Generation ×1 2D Spatial Locality ×1 Visual Token Prediction ×1 Document Filtering ×1 Disk-based Retrieval ×1 Multi-Shot ×1 Boundary-Aware ×1 AdaLN ×1 bounded noise ×1 linear contextual bandits ×1 set-membership estimation ×1 log regret ×1 optimism ×1 Misdirection ×1 Automated Attacks ×1 self-preference ×1 revision ×1 verifiable evaluation ×1 state checkpoint ×1 low-latency ×1 CUDA graph ×1 execution reuse ×1 Confident Decoding ×1 Entropy Valley ×1 Decoding Strategy ×1 Open Questions ×1 Contextual Privacy ×1 Annotation-Conditioned ×1 multimodal chain-of-thought ×1 perception tasks ×1 Predictive World Model ×1 Long-Horizon Planning ×1 Blocking Mechanism ×1 Object Counting ×1 Cycle-Consistency ×1 OS-level Agent ×1 Android ×1 Playlist Captioning ×1 Music Streaming ×1 Semantic Framing ×1 Multi-dimensional Evaluation ×1 Real-world Sessions ×1 Harness-Model Combination ×1 procedural memory ×1 skill transfer ×1 cross-model generalization ×1 CLI agents ×1 skill-mediated ×1 execution bottleneck ×1 plan decay ×1 replay pairing ×1 strict stripping ×1 Matryoshka embedding pooling ×1 Parametric Tool Retrieval ×1 Business Rule Grounding ×1 Knowledge Preservation ×1 Enterprise LLM ×1 VeriEvol ×1 Data Scaling ×1 Hypothesis-Test Verification ×1 premature commitment ×1 trajectory consistency ×1 Agency ×1 Autonomy ×1 GIC ×1 Agentive System ×1 outcome-only reward ×1 negative trajectory ×1 token-level reweighting ×1 student-teacher ratio ×1 Tapering ×1 Depth-aware ×1 rubric-gated compaction ×1 adaptive summarization ×1 KV-cache reuse ×1 conversational infill ×1 knowledge transfer ×1 input compression ×1 output compression ×1 accuracy ×1 semantic divergence ×1 Large-scale RecSys ×1 hash-constraint ×1 exhaustive-enumeration ×1 Task Routing ×1 Root Memory ×1 Implicit Logical Retrieval ×1 Personalized LLMs ×1 IMLogic ×1 intent-aware retrieval ×1 grocery e-commerce ×1 Sketch Supervision ×1 Online Data Mixing ×1 SAC ×1 Data Scheduling ×1 position-invariant ×1 low-rank patch ×1 LLM Annotation ×1 Distant Supervision ×1 Topological Structure ×1 Visual Hallucination ×1 Top-K Recommendation ×1 Semantic Proxy ×1 MNAR Bias ×1 Explainable Evaluation ×1 Autoregressive Modeling ×1 Contrastive-Free ×1 LLM Supervision ×1 Embedding Training ×1 Controlled Text Generation ×1 Counterfactual Policy Optimization ×1 Causal Consistency ×1 misinformation detection ×1 persistent memory ×1 Document Reasoning ×1 Workplace AI ×1 Language World Model ×1 Controllable Simulation ×1 KV-cache quantization ×1 bit allocation ×1 efficient serving ×1 attention logit error ×1 thinking tokens ×1 deliberation ×1 over-refusal ×1 AUROC prediction ×1 Decision History ×1 Practice Feedback ×1 set RL ×1 aggregation ×1 YaRN ×1 randomized positional encoding ×1 long-context reasoning ×1 OBS ×1 Personalized Generation ×1 Textual Gradient Descent ×1 Self-Confirmation Trap ×1 interactive AI ×1 Agentic Models ×1 Open Dataset ×1 Multi-Prefix Embedding ×1 Causal Language Models ×1 Jailbreak Detection ×1 Entropy Dynamics ×1 Intermediate Layer ×1 self-instruct ×1 PAVE ×1 Golden Set ×1 Brazilian Portuguese ×1 partial orders ×1 rank aggregation ×1 extreme classification ×1 zero-shot retrieval ×1 meta-classification ×1 large-scale retrieval ×1 Rademacher complexity ×1 agentic-ecommerce ×1 micropayments ×1 information-market ×1 trust ×1 NLP-challenges ×1 Conversational Product Search ×1 attribute-based preference elicitation ×1 patience modeling ×1 Amazon Reviews ×1 Variable-length Generation ×1 plasticity loss ×1 multilingual training ×1 Semantic Rewriting ×1 Data Sanitization ×1 belief representation ×1 amortized policy ×1 Bayesian experimental design ×1 Memory Recovery ×1 User State Reconstruction ×1 Noise Correction ×1 Autonomous Evaluation ×1 Cost-Performance ×1 iterative pruning ×1 Photography Guidance ×1 Capture-Time ×1 Composition ×1 Pose Recommendation ×1 Mathematics ×1 Graph Extraction ×1 Cross-Scenario ×1 Contrastive Evidence Gating ×1 Answer-Label-Free ×1 PACE ×1 exponential moving average ×1 LM fine-tuning ×1 Progress Advantage ×1 Bisimulation ×1 BitNet ×1 Multi-precision ×1 Temporal Perception ×1 Graphical Understanding ×1 Vision-Language Perception ×1 Paradigm Evolution ×1 Multimodal Intelligence ×1 Deterministic Anchors ×1 Repository Navigation ×1 Behavioral Stability ×1 multi-capability composition ×1 velocity field ×1 generative field distillation ×1 Session Embedding ×1 Clickstream ×1 Intent Taxonomy ×1 Financial Services ×1 RL collapse ×1 interleaved SFT-RL ×1 control token ×1 process supervision ×1 GPU Inverted Index ×1 Triton Kernels ×1 Scatter-Add ×1 Parallel Scoring ×1 RAG Evaluation ×1 Tree Drafting ×1 Causal Parallel Head ×1 vLLM Integration ×1 Memory Taxonomy ×1 User-centric Evaluation ×1 corpus poisoning ×1 token influence attribution ×1 utility-fairness trade-off ×1 scoring function ×1 post-processing ×1 semi-greedy ×1 cascaded inference ×1 cost-aware serving ×1 Tool Suppression ×1 Open-Weight LLM ×1 Grammar Masking ×1 Breadth Search ×1 Korean ×1 F1 evaluation ×1 Enumeration ×1 Research Map ×1 Citation Influence ×1 Agentic Image Generation ×1 Context Gap ×1 T2I ×1 Plan-Search-Memory ×1 IA-Bench ×1 Intent Segmentation ×1 Soft-TPP ×1 real-time voice ×1 emotional intelligence ×1 perception-action gap ×1 prosody ×1 IO-aware ×1 Hyperparameter Tuning ×1 Multiple Hypothesis Testing ×1 Statistical Guarantees ×1 E-values ×1 Video Generation Agents ×1 Reward Learning ×1 Visual Tokenization ×1 Native Resolution ×1 non-negative matrix factorization ×1 supervised topic model ×1 text-response modeling ×1 Model Cascade ×1 Relevance Annotation ×1 long reasoning ×1 Hallucination Cascade ×1 Verification Latency ×1 Delay Stability ×1 Corrector Placement ×1 Grounded Laplacian ×1 Prompt Composition ×1 Cross-Module Interference ×1 Behavioral Leakage ×1 Robust-TO ×1 Tool Orchestration ×1 Video Reasoning ×1 Multi-modal Agent ×1 Production ×1 thought experiment ×1 sparsity regularization ×1 monosemanticity ×1 vision foundation models ×1 conversational memory ×1 validity detection ×1 target-conditioned verification ×1 dual-evidence gate ×1 memory update ×1 non-destructive deployment ×1 denoising attention ×1 vision transformer ×1 subspace separation ×1 image classification ×1 Muon optimizer ×1 hierarchical matrix ×1 Incrementality ×1 Cannibalization ×1 Experiment Calibration ×1 multimodal-benchmark ×1 harmful-video ×1 content-moderation ×1 decoding methods ×1 sequence probability ×1 correctness alignment ×1 best-of-N ×1 Morphological Alignment ×1 MLLM efficiency ×1 information theory ×1 visual tokens ×1 gradient estimation ×1 model-assisted sampling ×1 stochastic optimization ×1 role-playing agents ×1 Agent Collaboration ×1 Progressive Search ×1 agentic BPM ×1 process harness ×1 policy-governed agents ×1 workflow automation ×1 TDF model ×1 decision-alignment ×1 prior-weighted utility ×1 UQ evaluation ×1 Semantic Job Search ×1 Hindsight Experience ×1 Autonomous Exploration ×1 Experiential Rules ×1 Interactive Decision Making ×1 flexible reasoning ×1 riddle ×1 cognitive bias ×1 strategy selection ×1 benchmark validity ×1 validation self-awareness ×1 building to the test ×1 software engineering ×1 Temporal Session-Based Recommendation ×1 Item Calibration ×1 Interval Scoring ×1 delta-rule ×1 content-aware-gating ×1 WY-form ×1 chunk-parallel ×1 memory-blind ×1 Norm Regularization ×1 Reward Alignment ×1 Architecture Evolution ×1 Silent Failure ×1 domain-specific QA ×1 quantitative reasoning ×1 hybrid pipeline ×1 nuclear engineering ×1 multimodal evaluation ×1 perception ×1 rubrics ×1 gated scoring ×1 Dialogue Analysis ×1 Collaborative Problem-Solving ×1 Metacognitive Regulation ×1 Open Relation Extraction ×1 Progressive Reasoning ×1 Comparative Reasoning ×1 Threshold-Free ×1 Adaptive Budget ×1 Bayesian bootstrap ×1 scalable approximation ×1 Dirichlet reweighting ×1 social media ×1 user profiling ×1 clarification ×1 ambiguity detection ×1 interactive QA ×1 multi-hop reasoning ×1 gradient-based attribution ×1 consistency rewards ×1 visual understanding ×1 entropy guidance ×1 feature steering ×1 look-ahead bias ×1 Temporal Graph Neural Networks ×1 Normalized Relevance Measure ×1 Information Flow ×1 Event-based Graphs ×1 Over-smoothing ×1 User-Item Decoupling ×1 Kuaishou ×1 Fine-Tuning Reversion ×1 Binary Decomposition ×1 reward_calibration ×1 score-based_optimization ×1 LLM_training ×1 group-relative_RL ×1 RiVER ×1 intent-aware training ×1 safety classification ×1 AIMS dataset ×1 faithful intent ×1 VLE fine-tuning ×1 WiSE-FT ×1 fashion retrieval ×1 multi-task contrastive ×1 co-failure ×1 β-ceiling ×1 multi-LLM ensemble ×1 voting ×1 mixture-of-agents ×1 Tandem Training ×1 Compatibility ×1 Multi-model Communication ×1 Personalized Landing Page ×1 Hierarchical Reward ×1 L-RTG ×1 dynamic sparsity ×1 gating network ×1 resource-adaptive inference ×1 layer skipping ×1 head pruning ×1 budget-conditioned ×1 multimodal fraud detection ×1 voice clustering ×1 rule-based scoring ×1 sampling strategy ×1 modal ceiling ×1 correlation ceiling ×1 effective sample size ×1 identifiability gap ×1 Gate Modulation ×1 Item Understanding ×1 Ontology Engineering ×1 S2D ×1 Zero-shot Recommendation ×1 Scholarly Recommendation ×1 Fault Detection ×1 Low-precision Attention ×1 Cross-task Generalization ×1 Text-to-Cypher ×1 Enterprise ×1 Revenue Maximization ×1 Permutation Algorithm ×1 Self-evaluation ×1 Attention Analysis ×1 LoRA Fine-tuning ×1 In-context QA ×1 Democratic ICAI ×1 Persona Debate ×1 modularity ×1 circuit analysis ×1 functional specialization ×1 cognitive domains ×1 hash-based LM ×1 vocabulary bottleneck ×1 autoregressive generation ×1 multilingual expansion ×1 Factual Error Detection ×1 Human-Written Text ×1 SMMD ×1 numerical prediction ×1 loss function ×1 kernel design ×1 listwise explanation ×1 semantic chunking ×1 multi-domain reasoning ×1 attention sorting ×1 lost-in-the-middle ×1 output-space allocation ×1 ROCKET ×1 perplexity-accuracy tradeoff ×1 masked diffusion model ×1 sample editing ×1 Sliding Window ×1 Negative Log-Likelihood ×1 LLM-Based Recommendation ×1 Recommendation Intuition ×1 BPR Loss ×1 Active Experimentation ×1 Multi-Agent Routing ×1 Set-Valued Prediction ×1 Supervised Router ×1 Agent Selection ×1 Training-Inference Mismatch ×1 FP8 Quantization ×1 Nugget Annotation ×1 Accountability ×1 Anchoring Bias ×1 computer-use-agent ×1 GUI-agent ×1 real-world-evaluation ×1 policy adherence ×1 sub-agent verifier ×1 dialogue-grounded ×1 safeguard ×1 Heuristics ×1 Semantic Spaces ×1 dataset distillation ×1 hyperbolic geometry ×1 low-rank alignment ×1 Unified Tokenization ×1 Dual-Stream Transformer ×1 1D Latent ×1 self-debate ×1 routing stability ×1 Advantage-aware Routing ×1 VLM Distillation ×1 Dual-Space Distillation ×1 Sinkhorn-Wasserstein ×1 Information Coverage ×1 Diverse Sub-queries ×1 next-state prediction ×1 multimodal learning ×1 context enrichment ×1 metadata ×1 model alignment ×1 processability hierarchy ×1 Activation Scaling ×1 Knowledge-action graph ×1 collaborative evaluation ×1 tool-augmented learning ×1 visual agent ×1 probe-based reward ×1 outcome gating ×1 generation ×1 InnerZoom ×1 Cross-layer Evidence ×1 Decoding Efficiency ×1 retriever optimization ×1 low-resource QA ×1 temperature learning ×1 query adaptation ×1 Supply Chain ×1 Vulnerability Scanning ×1 block size ×1 Incomplete Evidence ×1 Feature Splitting ×1 Feature Absorption ×1 Cross-sample Consistency ×1 Context Rot ×1 Long-horizon Search ×1 Search Drift ×1 Cross-request State ×1 Step-wise Reward ×1 Medical VQA ×1 Failure Cascades ×1 Data Referencing Errors ×1 Table Reasoning ×1 Social Platforms ×1 data construction ×1 text-rich image generation ×1 Multi-Block Decoding ×1 Teacher Forcing ×1 Block Buffer ×1 KV caching ×1 Image Synthesis ×1 hybrid-context ×1 packed sequences ×1 causal attention ×1 JIT compilation ×1 Probability Interface ×1 Multi-Instance ×1 Instance Masks ×1 token co-occurrence ×1 Cross-Space ×1 Bridge Module ×1 Latent Space Alignment ×1 Knowledge Transfer ×1 hierarchical memory ×1 quantitative retrieval ×1 long-term conversation ×1 in-memory architecture ×1 layer selection ×1 gating optimization ×1 Semantic BFS ×1 Mask-guided Search ×1 Partial-Order Learning ×1 Real-Time Sequence ×1 Ranking Signal Fusion ×1 Dual-Tower Retrieval ×1 Cascade Awareness ×1 Test-time Alignment ×1 Reward Decomposition ×1 Preference Realignment ×1 semantic compression ×1 zoom-in ×1 BiGRU ×1 multi-signal fusion ×1 RAGTruth ×1 Acceptance Criterion ×1 Tree Decoding ×1 asynchronous pipeline parallelism ×1 gradient delay ×1 optimizer robustness ×1 Error Feedback ×1 Retrieval Coverage ×1 Learned Hybrid Fusion ×1 Retrieve-then-Rerank ×1 Organizational Behavior ×1 Contextual Transaction Cost ×1 Workflow Design ×1 PI-FT ×1 Metadata Retrieval ×1 Embedding Fine-Tuning ×1 Field Dropout ×1 Cost-Efficiency ×1 Long-tail Knowledge ×1 MaaS ×1 Foresight ×1 Preference Tuning ×1 Attention Supervision ×1 Agent-Computer Interface ×1 CATE ×1 NCP-ToM ×1 Human Benchmark ×1 Context Decomposition ×1 Multiplicity ×1 decode routing ×1 PD-disaggregation ×1 expert-locality ×1 TPOT ×1 visual token pruning ×1 attention rectification ×1 entropy-guided ×1 MLLM inference ×1 Continuous Latent Reasoning ×1 Variational Learning ×1 Train-Inference Mismatch ×1 Fine-grained visual reasoning ×1 Vision-language model ×1 High-resolution ×1 PRA-GRPO ×1 multiprobe grid ×1 high-dimensional ×1 indexing cost ×1 LLM Training Recipe ×1 Homepage Construction ×1 Custom Tokenization ×1 Dr. GRPO ×1 Reward Standard Deviation ×1 Group-Size Law ×1 LLM ideation ×1 research taste ×1 distribution gap ×1 retrieval-heads ×1 non-literal-retrieval ×1 OV-circuit ×1 logit-contribution ×1 mechanistic-interpretability ×1 Semantic Cache ×1 Competitive Ratio ×1 Retrieval Buffer ×1 RLMF ×1 Browser Agent ×1 Natural Language Skills ×1 Skill Graph ×1 Structured Prediction ×1 Autoregressive Decoder ×1 Cognitive Neuroscience ×1 Functional Parcellation ×1 LLM Failures ×1 Brain Alignment ×1 Mechanistic Intervention ×1 Q-alignment ×1 Training-free Evaluation ×1 Orthonormal Initialization ×1 rule-based system ×1 symbolic reasoning ×1 rule generation ×1 Agentic Shopping ×1 Model-native Fulfillment ×1 OPD-RL ×1 Meta-reflection ×1 spoken language assessment ×1 ordinal regression ×1 layer routing ×1 Role-Typing ×1 Semantic Diversity ×1 Soft Query Refinement ×1 cross-modal hashing ×1 data-efficient ×1 stochastic neighborhood ×1 performance optimization ×1 runtime stability ×1 session-level retrieval ×1 co-occurrence clustering ×1 KB compression ×1 enterprise search ×1 coverage metric ×1 supervision on latents ×1 Benchmark Auditing ×1 State-Prediction Separation ×1 Graph Embedding ×1 Adaptive Masking ×1 Skill-based Evaluation ×1 LLM-based Task Generation ×1 bounded-memory ×1 typed-retrieval ×1 long-horizon-agent ×1 ablation-study ×1 Slay-the-Spire ×1 agent-evaluation ×1 metamemory ×1 deep learning theory ×1 approximation ×1 Attention Separation ×1 Training Acceleration ×1 Facet Disentanglement ×1 MIIT ×1 Multi-image ×1 Toxicity Detection ×1 Content Moderation ×1 Query Planning ×1 Logical Query Trees ×1 Exploratory Reasoning ×1 Thinking Mode ×1 Encoder-Free ×1 graph ANN index ×1 navigability signal ×1 tail recall ×1 local repair ×1 bursty churn ×1 Workflow Induction ×1 hard negative sampling ×1 IR Pipelines ×1 Experiment Plans ×1 PyTerrier ×1 Cascading Retrieval ×1 decision-aware training ×1 energy score ×1 cost-sensitive learning ×1 Block-Diffusion ×1 Re-ranker ×1 logarithmic ×1 4-bit ×1 Memory-induced Bias ×1 Parametric Absorption ×1 Instance-Specific ×1 Proxy Benchmark ×1 Instance Selection ×1 Regression ×1 LOOCV ×1 Emotional Feedback ×1 Program-as-Weights ×1 Fuzzy Functions ×1 Model Compilation ×1 Quasi-Monte Carlo ×1 arithmetic coding ×1 FVS ×1 low-selectivity ×1 recall-collapse ×1 adaptive-fallback ×1 in-filtering ×1 Training Steps ×1 Optimal Compute Allocation ×1 behavioral grounding ×1 QRI cards ×1 search evaluation ×1 inverse propensity scoring ×1 user preference alignment ×1 Agent Skill-Use ×1 Rubric Self-Evolution ×1 SFT Data Filtering ×1 reasoning-verification ×1 LLM-judge ×1 proof-trace ×1 structured-verification ×1 adversarial-robustness ×1 Distribution-wise Reward ×1 FID ×1 Model Weights ×1 Domain Proportions ×1 Weight Space ×1 adoption ×1 productivity ×1 organizational rollout ×1 CLI tools ×1 Pull Request ×1 3D Trajectories ×1 message passing ×1 preemption ×1 context efficiency ×1 personality adaptation ×1 behavior change ×1 fluid personality ×1 contextual calibration ×1 layer contribution ×1 transformer layer ×1 Evidence Packing ×1 Answer-in-Context ×1 Reader Budget ×1 open-world ×1 perturbation-augmented fine-tuning ×1 Zero-Shot CIR ×1 Proxy Task ×1 FoCo ×1 Zeroth-order optimization ×1 activation subspace ×1 memory-efficient fine-tuning ×1 quantized LLM ×1 Variational Bayesian ×1 Sparse Masking ×1 Retrieval-Augmented Reasoning ×1 Factual Error Correction ×1 Reasoning Language Models ×1 Knowledge Coherence ×1 Error Accumulation ×1 consistency loss ×1 AI tutoring ×1 cognitive scaffolding ×1 reasoning-centric design ×1 Multi-Agent Debate ×1 Social Structure ×1 Alignment Pressure ×1 Off-the-Record ×1 Emergent Objective ×1 modular generation ×1 decomposition ×1 Privileged Information Leakage ×1 Reverse-KL Barycenter ×1 Exploration Preservation ×1 Adaptive Distillation ×1 Attention Relevance ×1 Reasoning Drift ×1 Memory Injection ×1 Value Category ×1 Entropy-Aware ×1 Textual Noise ×1 Submodular Maximization ×1 Dense Instructions ×1 VLM Efficiency ×1 X-to-4D ×1 Gaussian Splatting ×1 Policy Evolution ×1 End-to-End Learning ×1 Hierarchical Landmark ×1 Label System ×1 matrix factorization ×1 prefix masking ×1 experience replay ×1 refusal subspace ×1 RFM-AGOP ×1 probe ×1 Neuron-Aware ×1 Active Few-Shot Learning ×1 Neuron Activation ×1 Dual-Criteria Selection ×1 Graph Spanner ×1 Approximate Nearest Neighbors ×1 Certify-Rectify ×1 speech language model ×1 AgentCanvas ×1 KDLoop ×1 Rollout Noise ×1 Behavioral Modeling ×1 Opinion Modeling ×1 PII ×1 Resurfacing Attacks ×1 Knowledge Removal ×1 LLM Persona ×1 Big Five ×1 Color Assignment ×1 Chart Choice ×1 Model Dependence ×1 Task Framing ×1 Online Monitoring ×1 Threshold Calibration ×1 Verifier Signal ×1 SAM Proposals ×1 Instance Separation ×1 Hadamard Rotation ×1 Lloyd-Max Codebook ×1 Data-Agnostic ×1 Long-horizon Prediction ×1 Self-review ×1 Practice-based Learning ×1 Fuzzing ×1 Skill Marketplace ×1 Implicit Intent ×1 Agent Self-evolution ×1 Minimal Pipeline ×1 Vibe Coding ×1 agentic code generation ×1 observational study ×1 Speech Control ×1 Interactive AI ×1 inference power prediction ×1 GPU efficiency ×1 sustainable AI ×1 regression model ×1 Typographic Attack ×1 semantic-chunking ×1 Self-Gating Attention ×1 Efficient Transformer ×1 Deployment-Oriented ×1 dialect steering ×1 sparse neurons ×1 Arabic LLMs ×1 History Utilization ×1 Decoder-Only ×1 relevance-based embeddings ×1 candidate retrieval ×1 heavy ranker ×1 support set selection ×1 CUR decomposition ×1 lightweight retrieval ×1 multi-modal reasoning ×1 Markov decision process ×1 vision-thinking guidance ×1 large-scale training ×1 convergence ×1 Dense Video Captioning ×1 Event-Factorized ×1 Latent Planning ×1 Action Evaluation ×1 Monte-Carlo Tree Search ×1 Q-value Distillation ×1 Trustworthiness ×1 Diagnostic ×1 Automated KG Construction ×1 mobile deployment ×1 Attention Dynamics ×1 Token-level Analysis ×1 Cross-modal Interaction ×1 Test-time Intervention ×1 Temporal Gap ×1 trajectory-based distillation ×1 candidate constraint ×1 LongEval-RAG ×1 evidence ranking ×1 sentence reranking ×1 Inventory-Aware ×1 Prompt Adaptation ×1 Research Ideation ×1 Pattern Mining ×1 Novelty Check ×1 relevance classification ×1 LLaMA2 ×1 Multi-Platform ×1 content-generation ×1 agent-skills ×1 poster-automation ×1 video-generation ×1 blog-generation ×1 tool orchestration ×1 Log-Sigmoid ×1 Continuous Feedback ×1 Eviction Policy ×1 Memory Compression ×1 Future Attention ×1 Online Pruning ×1 Length Bias ×1 Effective Information Length ×1 Reflexive ×1 LLM-as-a-Verifier ×1 Verification Scaling ×1 Trajectory Reward ×1 Fine-grained Feedback ×1 RL Reward Shaping ×1 VA-PMTP ×1 Task-Aware Tokens ×1 STARK ×1 Agentic RS ×1 Level of Autonomy ×1 Trajectory Review ×1 Production Benchmark ×1 Nightly Regression ×1 cannibalization ×1 personalized incentive ×1 SUTVA ×1 Token-level Loss ×1 buying agent ×1 Bayesian learning ×1 competitive ratio ×1 LLM-based agent ×1 Semi-Autoregressive ×1 Confidence Scheduling ×1 DeepSeek ×1 token merging ×1 vision-language retrieval ×1 ColPali ×1 object evidence ×1 storage efficiency ×1 Prefix Replay ×1 Multi-Turn Interaction ×1 Step-Decay ×1 Agent Routing ×1 Evaluation Validity ×1 Theory ×1 Signed MaxSim ×1 agent self-improvement ×1 meta-skill ×1 two-timescale ×1 claim graph ×1 long-form QA ×1 self-speculation ×1 tri-mode ×1 VLM Critic ×1 Unified Multimodal Generation ×1 Task Unification ×1 knowledge-distillation ×1 weak-to-strong ×1 Forward-KL ×1 turn-aware budgeting ×1 adaptive rollout ×1 KL supervision ×1 Reliability Control ×1 Digital Museum ×1 Length-Penalized RL ×1 Bias Hint ×1 Early Abort ×1 Probe Cascade ×1 Recall Control ×1 Inference Compute Saving ×1 Failure Prediction ×1 Deterministic Environment ×1 Stage-aware Strategy ×1 Dataset InPE ×1 video-to-music ×1 temporal reranking ×1 cross-modal embedding ×1 semantic compatibility ×1 Multi-Hop RAG ×1 Learnable Control ×1 State-Conditioned Policy ×1 Evidence Operations ×1 Sufficiency Feedback ×1 modality interference ×1 gradient conflict ×1 full-duplex SLM ×1 hierarchical parameter separation ×1 semantic coherence ×1 speech-language model ×1 KOL matching ×1 multi-stage cascade ×1 SimPO ×1 pointwise scoring ×1 Context Extension ×1 set-level compatibility ×1 efficient set exploration ×1 Kronecker-Factored Hessian ×1 Bidirectional Incoherence ×1 2-bit Quantization ×1 Extragradient ×1 SAM ×1 Long-Context Understanding ×1 Instruction Synthesis ×1 Lost in the Middle ×1 CoT Compression ×1 Attention-based Scoring ×1 Student-Aware ×1 depth sharing ×1 frequency decomposition ×1 Metadata Generation ×1 RDF ×1 BPMN ×1 Process Modeling ×1 multi-vector ×1 token pooling ×1 Document Context ×1 Asymmetric Optimization ×1 Interaction Context ×1 Real-Time Generation ×1 Causal Pretraining ×1 Sparse Delta Memory ×1 Product-Key Memory ×1 IsoFLOP ×1 DeltaNet ×1 cross-layer routing ×1 Embedding Retrieval ×1 LLM Navigation ×1 Agon ×1 Competitive RL ×1 Data-Science Agent ×1 Terminal Benchmark ×1 Sub-task Decomposition ×1 RLVP ×1 Path Penalty ×1 Agent Deployment ×1 Constraint Satisfaction ×1 Dense Prediction ×1 Readout ×1 Flux ×1 SOP ×1 Causal Extraction ×1 Multi-Agent Search ×1 Role Factorization ×1 Capacity Allocation ×1 confidence distillation ×1 pre-solution ×1 Asynchronous RL ×1 Tool-Use Calibration ×1 Behavior Imbalance ×1 Token-Level Divergence ×1 unlearning ×1 feedback regularization ×1 Safety Gymnasium ×1 Motion Generation ×1 Emotion Expression ×1 Animation ×1 Function Composition ×1 Attribute-Guided Reflection ×1 activation dispersion ×1 entity familiarity ×1 factual reliability ×1 event-based vision ×1 video diffusion models ×1 autoregressive unrolling ×1 frame interpolation ×1 zero-shot generalization ×1 sensor robustness ×1 Logical Query Answering ×1 Graph Wavelets ×1 Multi-Hop Query ×1 EFO Queries ×1 LMLM ×1 Continuous Query ×1 Factual Precision ×1 Perplexity ×1 Panoramic Generation ×1 Depth Generation ×1 Circular Padding ×1 Similarity Loss ×1 Chain-of-Frame ×1 Reasoning Tokens ×1 Bellman Equation ×1 Stochastic Approximation ×1 Convergence Analysis ×1 Compliance Rectification ×1 Curriculum Reinforcement Learning ×1 Group‑Relative Experience ×1 Video Ad ×1 Rejection Fine-tuning ×1 Symbolic Reasoning ×1 Time-Series Query ×1 Natural Language Interface ×1 Sketch-based Query ×1 LLM Parsing ×1 Frequency Usage ×1 Data-Driven ×1 failure localization ×1 agent attribution ×1 confidence-aware aggregation ×1 Source Grounding ×1 Knowledge Infrastructure ×1 Quality Gate ×1 Backtracking ×1 Validation-guided ×1 Correction-aware ×1 Attribute Extraction ×1 Data Validation ×1 Capability-Driven ×1 Closed-Loop Evaluation ×1 social norms ×1 human-AI coordination ×1 dynamic interaction ×1 reward design ×1 representation theory ×1 sinusoidal filters ×1 whitening ×1 waterfilling ×1 stationary statistics ×1 post hoc linearization ×1 sink tokens ×1 fixed-budget cache ×1 transformer inference ×1 Prefix Control ×1 persona evolution ×1 self-locking ×1 OSO loop ×1 context gravity ×1 divergence targeting ×1 Mixture Model ×1 Power-Spherical ×1 Admixture ×1 Spherical Coding ×1 Codebook-Free ×1 on-device deployment ×1 text summarization ×1 Sequential Testing ×1 Confidence Interval ×1 Monte Carlo training ×1 gradient-free optimization ×1 deep neural networks ×1 discrete weights ×1 unconventional transfer functions ×1 evaluation reliability ×1 jury aggregation ×1 Draft-and-Verify ×1 Hybrid Sparse-Dense ×1 Visual Updates ×1 TSIM Router ×1 StructCoT ×1 ULMM ×1 hierarchical document ×1 structure-aware indexing ×1 document filtering ×1 Domino ×1 conditional scoring ×1 GPU-native builder ×1 Adaptive Generation Order ×1 Edge Differential Privacy ×1 Jaccard Similarity ×1 semantic persistence ×1 knowledge objects ×1 derive vs infer ×1 conceptual model ×1 memory agent ×1 proactive intervention ×1 behavioral state decay ×1 document hashing ×1 deduplication ×1 embedding-based hashing ×1 OCR post-correction ×1 historical documents ×1 multilingual benchmark ×1 statistical test ×1 policy evaluation ×1 heterogeneous treatment effects ×1 recommendation systems ×1 UMAP ×1 kNN graph ×1 k-core decomposition ×1 clustering coefficient ×1 sensemaking ×1 multimodal interpretability ×1 concept consistency ×1 structured sparsity ×1 Synthetic Speech ×1 Low-Resource ×1 visual pretraining ×1 document understanding ×1 Knowing-Using Gap ×1 self-patching ×1 knowledge-circuit misalignment ×1 Knowledge Modeling ×1 Document Analysis ×1 correctness agreement ×1 attention analysis ×1 behavioral divergence ×1 Super Weights ×1 Sparse Training ×1 Intrinsic Dimensionality ×1 answerability ×1 hidden state probes ×1 Log Analysis ×1 Symbolic Compression ×1 Incident Diagnosis ×1 Microservices ×1 Sovereign AI ×1 German NLP ×1 Procedural Similarity ×1 Repository-Level Code Generation ×1 Static Analysis Feedback ×1 Rubric Evaluation ×1 Episodic Visual Memory ×1 Long-horizon Dialogue ×1 Agentic Inequality ×1 Context Access Divide ×1 cognitive load ×1 knowledge workers ×1 Generative AI ×1 Learning ×1 Randomized Experiment ×1 Augmentation vs Automation ×1 Knowledge Retention ×1 Education ×1 Hoyer ×1 nuclear norm ×1 Agent Workload ×1 Session Scheduling ×1 Runtime Auditing ×1 Semantic Firewall ×1 Token Flow ×1 activation aggregation ×1 cross-layer injection ×1 Conversational IR ×1 Ad-hoc Search ×1 Data Refinement ×1 Programmatic Editing ×1 recursive-delegation ×1 web-search ×1 LLM-agent ×1 information-seeking ×1 relaxed verification ×1 draft model ×1 speed-quality trade-off ×1 Training Strategies ×1 Cosine Annealing ×1 Cyclic LR ×1 budget-aware ×1 resampling ×1 rerouting ×1 backdoor-attack ×1 code-completion ×1 forensic ×1 data-poisoning ×1 market stability ×1 mediation ×1 social dilemma ×1 Self-Guided ×1 FactVerification ×1 TwoStageTraining ×1 Auction Mechanism ×1 Sparse Fine-Tuning ×1 Activation-Aware Pruning ×1 Wanda ×1 exploration_collapse ×1 Riemannian_manifold ×1 policy_optimization ×1 Process-Oriented ×1 Empirical Study ×1 Raw IDs ×1 Multi-stakeholder ×1 Recommender System Evolution ×1 Segmentation ×1 VIP-SAM ×1 Domain Knowledge ×1 Unstructured Data ×1 Latent Factors ×1 NRFR ×1 Rashomon Explanation ×1 Interpretable ML ×1 Double-loop Learning ×1 Agentic Context ×1 Persistent Instruction ×1 Typed Semantic Graph ×1 Local Verification ×1 GRACE ×1 time-series-forecasting ×1 linear-models ×1 gated-routing ×1 Retrieval Policy ×1 Search Granularity ×1 Tool Coordination ×1 Task-Oriented Dialogue ×1 Constraint Satisfaction Problem ×1 Consistency Detection ×1 gradient_boosting ×1 hidden_confounding ×1 spectral_deconfounding ×1 early_stopping ×1 mixed_models ×1 Scene Description ×1 CSB Dataset ×1 Visual Document Understanding ×1 Distribution Alignment ×1 Evidence Integration ×1 multilingual reasoning ×1 parseability ×1 process reward model ×1 Realignment ×1 Fraud Detection ×1 Semantic State ×1 DQN ×1 selective persistence ×1 zero-token refresh ×1 collaborative workspace ×1 Dynamic Token Selection ×1 Foveation ×1 Human Visual System ×1 MultiModalMemory ×1 SelfEvolution ×1 EmbodiedAI ×1 Predictive Divergence Mask ×1 Top-K Estimation ×1 Medical Imaging ×1 automated research ×1 design patterns ×1 generative taste ×1 novel view synthesis ×1 asymmetric architecture ×1 long-sequence ×1 neural rendering ×1 Tool-Adaptive ×1 Cost-Aware RL ×1 Pointwise Scoring ×1 Egocentric AI ×1 Personal Assistant ×1 Streaming Processing ×1 Aligned Recommendation ×1 Label Shift ×1 Normative Design ×1 Attribute Distribution ×1 User Engagement ×1 continued training ×1 stochastic convex optimization ×1 horizon-free ×1 Warmup-Stable-Decay ×1 Side Adaptation ×1 Stream-aware Fusion ×1 Residual Adapter ×1 Large Pre-trained Models ×1 proof generation ×1 proof verification ×1 automatic evaluation ×1 multimodal generation ×1 cross-modal coupling ×1 joint understanding ×1 bilingual ×1 low-cost training ×1 Pattern Matching ×1 Unsupervised Ranking ×1 Document Identifier ×1 Business-value Ranking ×1 Collision-free ×1 score-only distillation ×1 PairMSE loss ×1 View Augmentation ×1 Student-Guided Perturbation ×1 Cyclic Shift ×1 Assortment Optimization ×1 Submodularity ×1 MNL ×1 Peer Selection ×1 Behavior Localization ×1 Code Understanding ×1 LLM-Assisted ×1 Proxy Exploration ×1 Update Signal Transfer ×1 ExplainableAI ×1 ConversationalScams ×1 UserStudy ×1 inference scaling ×1 external feedback ×1 proposer-reviewer ×1 Selective Query Expansion ×1 Participatory Audit ×1 Vision Language Models ×1 offline-evaluation ×1 popularity-bias ×1 preference-induction ×1 pooling ×1 top-N-recommendation ×1 Web Design ×1 Agent Usability ×1 Decision Reliability ×1 Streaming Inference ×1 visual tool-calling ×1 Document AI ×1 Visual Pretraining ×1 Pretraining Strategy ×1 On-Device Agent ×1 Device Tools ×1 Execution Boundaries ×1 Configuration-Driven ×1 Knowledge Graph Construction ×1 Deduplication ×1 Community Detection ×1 Entity-Oriented Memory ×1 Video Stream ×1 Zero RL ×1 Emergent Reasoning ×1 1T Parameters ×1 Agent safety ×1 Vulnerability concept graph ×1 Auto-research ×1 Falsifiable loop ×1 Production agents ×1 Repetitive Suffixes ×1 Short-to-long Schedule ×1 Associative Memory ×1 Recurrent Memory ×1 Hourglass Reasoning ×1 Symbolic Representation ×1 MisleadingChartQA ×1 Cross-modalVerification ×1 Spatio-Temporal Grounding ×1 Evidence-Backed ×1 Video LLM ×1 Offline-to-Online RL ×1 Policy Selection ×1 Performance Forecast ×1 Email Communication ×1 Tone ×1 Mediation Analysis ×1 Emotional Positivity ×1 complexity-aware ×1 execution planning ×1 cost reduction ×1 llm ×1 NTP ×1 对比学习 ×1 跨域学习 ×1 训练信号 ×1 序列推荐 ×1 Generation ×1 One-Class Learning ×1 Neural CDE ×1 Debugging ×1 Recursive Reasoning ×1 Interest Decomposition ×1 Deep Supervision ×1 Inference-Time Computation ×1 Latent Control ×1 Agent Decision Making ×1 Deployment-time Control ×1 Multi-Domain Retrieval ×1 Contamination Control ×1 personalized generation ×1 highlight retrieval ×1 video thumbnail ×1 VLM-guided ×1 click engagement ×1 Discount-Aware ×1 Fourier Transform ×1 Fill-in-the-Middle ×1 Function Call ×1 Keyframe-Conditioned Video Generation ×1 Multi-Modal ×1 Video Quality ×1 Keyframe Execution ×1 LLM Hallucination ×1 Lean Theorem Prover ×1 Audit Trail ×1 OpenClaw ×1 Tucker decomposition ×1 Johnson-Lindenstrauss ×1 variance components ×1 multiverse analysis ×1 long-horizon conversations ×1 memory operations ×1 scaling bias ×1 gated bias correction ×1 fresh item retrieval ×1 Vision-Free ×1 Attribute Augmentation ×1 Instagram ×1 embodied cognition ×1 multimodal planning ×1 visual imagination ×1 Mixture-of-Transformers ×1 robot action generation ×1 Offline Replay ×1 Retrieval Optimization ×1 DeepNorm ×1 Depth Scaling ×1 metric co-evolution ×1 evolutionary lifecycle ×1 anchor discipline ×1 Representation Anchoring ×1 Language-Action Alignment ×1 representation autoencoder ×1 discrete tokens ×1 alignment distillation ×1 Vietnamese ×1 Hotel Recommendation ×1 Quality Control ×1 video LLM ×1 visual dependency ×1 language priors ×1 black-screen baseline ×1 Interest Exit ×1 Post-purchase Redundancy ×1 Dual-path Cross-Attention ×1 Satiation Mechanism ×1 Self-supervised Learning ×1 inference-optimization ×1 byte-exact ×1 knowledge-flywheel ×1 frozen-model ×1 Consistency Regularization ×1 Image-Text Mismatch ×1 Attribute Clustering ×1 Representation Space ×1 Contrastive Policy Optimization ×1 Token-level Correctness ×1 Entropy Alternative ×1 agent-optimization ×1 continual-learning ×1 harness-optimization ×1 regression-aware ×1 prompt-engineering ×1 Evidence Reliability ×1 Multi-step QA ×1 cross-device agents ×1 heterogeneous environments ×1 task evaluation ×1 multi-device collaboration ×1 Skill Wiki ×1 Resource Distillation ×1 Software Agents ×1 Incremental Search ×1 Hybrid Embeddings ×1 Label-Free Self-Distillation ×1 Consensus Anchoring ×1 Grounded Reasoning ×1 Statistical Self-Consistency ×1 Probabilistic Inference ×1 Macro Fallacy ×1 Backend-Agnostic ×1 index partitioning ×1 probing function ×1 auction algorithm ×1 heavy-tailed ×1 random clock ×1 logsignature ×1 tail control ×1 Length Exploitation ×1 Signal Regulation ×1 Exploration Catalyst ×1 Advantage Clipping ×1 One-Shot Learning ×1 Citation Integrity ×1 Scientific QA ×1 Hyper-Connections ×1 Residual Stream Expansion ×1 Sparse Architecture ×1 Historical Analogies ×1 Structural Decomposition ×1 Cross-Analogy Confirmation ×1 Foresight Analysis ×1 MeanFlow ×1 Forward-Process RL ×1 Graph-Based Framework ×1 Composable Systems ×1 Multi-agent system ×1 Search agent ×1 Stateful orchestration ×1 Relational schema completion ×1 Evidence graph ×1 Pipeline parallelism ×1 Citation Optimization ×1 Search Visibility ×1 Evidence Hierarchy ×1 video-encoding ×1 world-event-decomposition ×1 real-time-interaction ×1 pretraining-task ×1 Feature Heterogeneity ×1 Signal Dilution ×1 Decoupled Architecture ×1 Downstream Rewards ×1 User Retention ×1 Proxy Rewards ×1 Model-agnostic ×1 Long-term Engagement ×1 Data-efficient learning ×1 Attention Head Reweighting ×1 Few-shot classification ×1 Temporal Difference ×1 Video MLLM ×1 I3D-ViT ×1 controllability ×1 black-box auditing ×1 position-aware confidence ×1 agent learning ×1 hindsight skills ×1 Position Debiasing ×1 Significance Test ×1 Pathology-Keyed Archive ×1 Autonomous Commerce ×1 Loyalty ×1 Human-Agent Alignment ×1 Blockchain ×1 exemplar memory ×1 audio-image-text ×1 Prior Fusion ×1 Efficient Proxy ×1 Dynamic Fusion ×1 Rubric Evolution ×1 Pairwise Preference ×1 Query-Only ×1 No Annotation ×1 Coordinated-Behavior ×1 Feedback-Loop ×1 Agent-Based-Simulation ×1 Recommendation-Robustness ×1 Offline-Evaluation ×1 neuro-symbolic ×1 Z3 solver ×1 educational QA ×1 data science ×1 transition prediction ×1 cost-aware routing ×1 video DiT ×1 Off-policy evaluation ×1 kernel methods ×1 bias-variance trade-off ×1 LLM Re-Ranker ×1 Interactive Training ×1 Control Plane ×1 Training Steering ×1 Counterfactual Utility ×1 Bridge Evidence ×1 Entity Discrimination ×1 Mutable Sketches ×1 KP-tree ×1 Low-rank Projection ×1 Embedding Freshness ×1 variable-rate ×1 math ×1 cost-effective ×1 Activation Checkpointing ×1 Model Architecture ×1 Layer-loop ×1 Compute-matched scaling ×1 SPT ×1 RL Reasoning ×1 audio-video editing ×1 multi-modal agent ×1 cross-modal consistency ×1 Human-Alignment ×1 Persona-Based Testing ×1 Reliability-Gating ×1 Contextual Rubrics ×1 Plan-Centric Interaction ×1 Interactive Task Automation ×1 Memory Hub ×1 Hybrid-modal ×1 RLRF ×1 Text-to-SPARQL ×1 Schema-Aware ×1 scene representation ×1 spatial grounding ×1 semantic-spatial binding ×1 Next-State Prediction ×1 XMLC ×1 Subject Indexing ×1 Multi-label Classification ×1 Systematic Misalignment ×1 Diagnosis ×1 Model Auditing ×1 Middle-Layer Recurrence ×1 mask-aware ×1 masked generation ×1 Hierarchical Global Attention ×1 VRAM-efficient ×1 tokenizer expansion ×1 embedding initialization ×1 continued pre-training ×1 Delta Signal ×1 Rectified Flow ×1 Ridge Regularization ×1 Generalized Cross-Validation ×1 Active Vision ×1 MLLM Evaluation ×1 Perception-Reasoning Loop ×1 Bayesian prompt optimization ×1 Gradient-guided discrete sampling ×1 Parallel tempering ×1 Prompt optimization ×1 Hyperparameter Sensitivity ×1 Shapley Effects ×1 Multi-Objective Learning ×1 Coverage-Adaptive ×1 Self-explanation ×1 Plausibility ×1 Actionability ×1 CRAFT ×1 capability diagnosis ×1 fine-tuning data ×1 rubric clustering ×1 Task Vectors ×1 Data Pipeline ×1 Grounded Platform ×1 Single-Agent System ×1 Communication Compression ×1 Model Capability ×1 entropy control ×1 group entropy ×1 exploration-exploitation ×1 bootstrapping ×1 Tree Rollout ×1 Dialogue State Tracking ×1 preference modeling ×1 Grounded Video QA ×1 Self-Correcting Agent ×1 Hierarchical Search ×1 Adaptive Decoding ×1 Lookahead Decoding ×1 relevance decoding ×1 LLM judgment ×1 LLM Simulation ×1 API Calling ×1 GiGPO ×1 learning rate ×1 2D-RoPE ×1 Copy Task ×1 Short-Video Rec ×1 Feedback Cleaning ×1 Skill Curation ×1 SkillCorpus ×1 vector quantization ×1 codebook collapse ×1 distribution matching ×1 Prospective Hypothesis Discovery ×1 Arena Evaluation ×1 Retrospective Context Regression ×1 Test-time Reasoning ×1 Induction Heads ×1 Bidirectional Circuit ×1 Influence Matching ×1 Outcome Alignment ×1 Sample-Level Influence ×1 Vision-Language Distillation ×1 citation analysis ×1 propositional reasoning ×1 LLM multi-agent systems ×1 surrogate models ×1 behavioural cloning ×1 mean-field theory ×1 agent-based modelling ×1 Trie Optimization ×1 Term ID ×1 Minimum Set Cover ×1 Distilled RL ×1 Cross-Family Distillation ×1 Satisfaction Label Bias ×1 Pairwise Learning ×1 Short Video Recommendation ×1 sparse embeddings ×1 content-based recommendation ×1 MERC ×1 Emotional Inertia ×1 Conversation ×1 agent retrieval ×1 capability profiles ×1 text embedding ×1 model fine-tuning ×1 kernel-methods ×1 associative-recall ×1 throughput-optimization ×1 Triton-kernels ×1 Subject-driven ×1 Attention Alignment ×1 Variance Decomposition ×1 Verifier-Guided Repair ×1 Non-Verifiable Tasks ×1 Reward Hacking Mitigation ×1 Latent Time Representation ×1 Activation Geometry ×1 scientific diagram generation ×1 structure-faithful ×1 multidimensional evaluation ×1 Wukong ×1 HSTU ×1 Scalable RecSys ×1 agent debugging ×1 DeepDebug ×1 recovery loop ×1 observability ×1 Group Decision Making ×1 Persona-based Discussion ×1 Travel Planning ×1 graph-based ×1 modular design ×1 filtered search ×1 dynamic updates ×1 C++ library ×1 Indexing Optimization ×1 CRUMB ×1 Reuse ×1 staleness adaptive ×1 MoE routing ×1 semantic uncertainty ×1 posterior inference ×1 stability under paraphrasing ×1 reasoning control ×1 self-looping ×1 latent state transition ×1 Multimodal Humor ×1 Humor Generation ×1 Financial QA Benchmark ×1 Complex Layouts ×1 Agent-as-a-Judge ×1 Corpus Alignment ×1 Feature-Gated Reranker ×1 SEC Filing ×1 Fact-checking ×1 listwise reranker ×1 structured retrieval ×1 hallucination containment ×1 calibrated abstention ×1 evidence-based confidence ×1 Compact Models ×1 Counterfactual Advantage ×1 efficient-generation ×1 rectified-flow ×1 latent-tokenizer ×1 image-editing ×1 model-distillation ×1 diffusion-transformer ×1 Counterfactual Decomposition ×1 Dynamic Top-K ×1 Agent Planning ×1 optimizer state ×1 SkewAdam ×1 tiered allocation ×1 caption quality ×1 data distribution ×1 procedural testbed ×1 Hypencoder ×1 Matryoshka Embeddings ×1 Q-Net ×1 HyperNetwork ×1 Trade-off ×1 Privacy Preserving ×1 Dual-tower ×1 Output Masking ×1 Token-level Advantage ×1 Factual Completeness ×1 Long-form Evaluation ×1 Meta-Rubric ×1 hybrid RAG ×1 trace-driven ×1 policy-guided ×1 Weighted Averaging ×1 Dimensional Adaptation ×1 Parameter Interpolation ×1 Graph Low-pass Filter ×1 Document Operations ×1 Verifiable Evaluation ×1 Long-range Tasks ×1 multi-head attention residuals ×1 depth routing ×1 forced compromise ×1 compute-equivalent gain ×1 contextual entrainment ×1 conversational search ×1 set cover ×1 answer reconstruction ×1 search density ×1 Isospectral Optimization ×1 Spectral Inheritance ×1 Multi-vector retrieval ×1 Centroid ×1 Exposure ×1 Baseline Corrections ×1 Marginalization ×1 Auto-Differentiation ×1 Evolving Intent ×1 Multi-turn Evaluation ×1 Simulation Framework ×1 Dynamic User Intent ×1 rubric-based ×1 PyTorch ×1 TrainingFramework ×1 NVIDIA ×1 requirement-driven ×1 data scaling ×1 Object-Oriented Programming ×1 Python ×1 post-hoc explainability ×1 spectral biclustering ×1 block-deletion ×1 model-agnostic ×1 click model ×1 evidence deep learning ×1 beta distribution ×1 Ascend NPU ×1 MFU ×1 surrogate policy ×1 outcome-reward RL ×1 Co-speech Gesture ×1 Digital Human ×1 Scene Grammar ×1 Business Value ×1 Query-aware SID ×1 Value-aware Ranking ×1 safety benchmark ×1 evidential ceiling ×1 rare harm ×1 hypothesis testing ×1 Recursive Agent ×1 Context Update ×1 Clarification Need ×1 Mixed Initiative ×1 Efficient Clustering ×1 Set Cover ×1 Contamination Resistance ×1 Open Evaluation ×1 Sandboxed Scoring ×1 Multi-turn Retrieval ×1 Diverse Intents ×1 DIM-Fashion ×1 Scene Graph ×1 Diffusion Model ×1 LLM interaction ×1 configurable agents ×1 experimental control ×1 behavioral logging ×1 user study platform ×1 Project Building ×1 Heterogeneous Graphs ×1 Cardinality-Decomposed Loss ×1 Cross Entropy ×1 User Reactivation ×1 Gap-Driven Gating ×1 Dimensional Routing ×1 Asymptotic Normality ×1 full-ranking ×1 reflection mechanism ×1 State Space Models ×1 User Embeddings ×1 Transactional Sequences ×1 Curriculum Cognition ×1 Educational LLMs ×1 Dynamic Compute ×1 Efficiency-Accuracy Trade-off ×1 Third-Party Skills ×1 Lead Ranking ×1 Listwise Optimization ×1 Discriminative Model ×1 Misleading Knowledge ×1 False Conclusions ×1 Korean LLM ×1 Branch Packing ×1 visual evaluation ×1 agentic protocol ×1 Bayesian online learning ×1 expert aggregation ×1 adaptive prediction ×1 conformal inference ×1 Gaussian processes ×1 Compaction ×1 Capacity ×1 Compression-Measurement ×1 Explanation ×1 Necessity ×1 Causal ×1 window attention ×1 everyday knowledge ×1 popular culture ×1 usability ×1 ranking explanation ×1 think-aloud study ×1 Neuro-Symbolic AI ×1 Deterministic Reasoning ×1 Rule Enforcement ×1 Rotary Positional Encoding ×1 Anti-Periodic ×1 Needle-in-a-Haystack ×1 Seed Lottery ×1 Cross-Component ×1 Evaluation Feedback ×1 Quality-Diversity Search ×1 Yield Metric ×1 Lineage-Based Evolution ×1 Probabilistic Residual Learning ×1 Causal Debiasing ×1 Skill-SP ×1 Product recommendation ×1 KV-cache eviction ×1 randomized design ×1 error certificate ×1 token-level detection ×1 LLM-generated text ×1 kernel smoothing ×1 Lepski's method ×1 human-AI coauthoring ×1 progressive cramming ×1 transformer capacity ×1 attention knockout ×1 reconstruction ×1 controllable recommendations ×1 text bottleneck ×1 content-based ×1 Kronecker ×1 TensorCore ×1 seed pruning ×1 particle filtering ×1 flow models ×1 variable-length generation ×1 discrete flows ×1 Multilingual Reranker ×1 Language Coherence ×1 Sparse Feature Ranking ×1 Embedding Tables ×1 Modular Framework ×1 Heterogeneous Memory ×1 Hierarchical Decoding ×1 Item Context-Aware Attention ×1 Path Reranking ×1 Dual-Path Decoding ×1 EVP ×1 FLAS ×1 graph-indexing ×1 cache-locality ×1 federated learning ×1 prompt tuning ×1 data heterogeneity ×1 low-rank decomposition ×1 Iterative Refinement ×1 Stochastic Tokenizer ×1 Multi-modal Reasoning ×1 Open-ended Generation ×1 Russian NLP ×1 Context-Weighted Training ×1 Perplexity Reduction ×1 CTMC ×1 VCSD ×1 contrastive distillation ×1 implicit reasoning ×1 self-reconstruction ×1 next-token prediction ×1 LLM retriever ×1 StateAct ×1 code-first agent ×1 multi-agent system ×1 state-grounding ×1 Feedback-guided Generation ×1 Early Detection ×1 Hilbert Space ×1 Neuron Merging ×1 Low-Rank Projection ×1 Persona Subspace ×1 multimodal pre-training ×1 Depth Sparsity ×1 Adaptive Token Retention ×1 Cosine Similarity ×1 Alignment Training ×1 LLM-generated text detection ×1 wavelet transform ×1 voting ensemble ×1 token log-probability ×1 Green Recommendation ×1 Differentiable Ranking ×1 Gradient Projection ×1 task conflicts ×1 on-policy training ×1 advantage decay ×1 hardware-software co-design ×1 importance-aware pruning ×1 element-level sparsity ×1 chunk-based sorting ×1 J-CoT ×1 J-space ×1 recurrent reasoning ×1 vocabulary-indexed coordinates ×1 Intent Discovery ×1 Semantic Propagation ×1 Contrastive Reasoning ×1 Graph Curvature ×1 Oversmoothing ×1 Oversquashing ×1 Entropic Curvature ×1 GNN Architecture ×1 Graph Rewiring ×1 AgentRCA ×1 Digital Twin ×1 Evidence-Grounded Reasoning ×1 Byte-Prefix Marginalization ×1 condition number ×1 efficient fine-tuning ×1 grounding displacement ×1 skill-description osmosis ×1 verification displacement ×1 paired evaluation ×1 capability domains ×1 subskills ×1 coverage audit ×1 code agent ×1 office agent ×1 compact model ×1 full-funnel ×1 industrial-scale ×1 request state ×1 session-level evaluation ×1 AI search ×1 conversation analysis ×1 novelty ×1 contextual insensitivity ×1 model-specific standards ×1 Interactive Bending ×1 Stable Diffusion ×1 Creative Practice ×1 ComfyUI ×1 Longitudinal Evaluation ×1 Architecture Crossover ×1 Veracium ×1 token probability ×1 image degradation ×1 error detection ×1 ranking manipulation ×1 protocol validity ×1 benchmark auditing ×1 HackDetect ×1 score inflation ×1 open-ended tasks ×1 reward engineering ×1 Creative Agents ×1 Canvas-Native ×1 semantic materialization ×1 sparse serving ×1 donor pair ×1 Claw-native recommendation ×1 user state modeling ×1 task-aware ×1 marginal curation ×1 creative optimization ×1 adaptive testing ×1 ranking model ×1 inference-time critic ×1 implicit associations ×1 codec ×1 efficient VLM ×1 dual-system ×1 intruder dimensions ×1 singular vectors ×1 Manifold Regularization ×1 Protocol Distillation ×1 Style Drift ×1 Relevance-Aware Search ×1 Ripgrep ×1 Coarse-to-Fine Guidance ×1 Search Convergence ×1 context utilization ×1 long-horizon recommendation ×1 semantic atoms ×1 propensity weighting ×1 conservative offline RL ×1 exposure bias ×1 LLM-as-compiler ×1 Game Solving ×1 Turn-Level Teaching ×1 Auto-Bidding ×1 Context Learning ×1 Fast-and-Slow Reasoning ×1 SID Routing ×1 Structure-grounded Reasoning ×1 conditional alignment ×1 streaming training ×1 unbiased data ×1 cascaded binary classification ×1 step-level GRPO ×1 relevance estimation ×1 token-level credit assignment ×1 rubric-conditioned RL ×1 language model training ×1 Hidden Activations ×1 Singular Values ×1 Spectral Methods ×1 Hybrid Memory ×1 text-space optimization ×1 rubric generation ×1 decoupled objectives ×1 Compositional Generalization ×1 prompt-injection ×1 adversarial-training ×1 Training-Serving Consistency ×1 Embedding Cache ×1 Online Training ×1 Memory Layer ×1 MPZCH ×1 RL-tuning ×1 Step-level Alignment ×1 Personalized Exploration ×1 information flow control ×1 taint tracking ×1 context branching ×1 declassification ×1 Temporal User Modeling ×1 Preference Evolution ×1 any-to-any ×1 Specialist Fleet ×1 Redundancy Loss ×1 Hybrid Serving ×1 skill-driven ×1 audio-video ×1 Modality Decoupling ×1 Cross-Modal Salience ×1 Behavior-Conditioned Generation ×1 Entropy-Aware Self-Distillation ×1 MoE Deployment ×1 Contrastive Distillation ×1 visual prompt engineering ×1 video reasoning ×1 safety classifier ×1 policy-adaptive ×1 binary QA ×1 content moderation ×1 Attention Mechanism ×1 Decoder-Only Transformer ×1 Prefix-Causal Attention ×1 revision loops ×1 stopping policies ×1 long-context decoding ×1 page-level summarization ×1 vLLM plugin ×1 Item-Item Graph ×1 Triadic Closure ×1 BPR Augmentation ×1 Block Diffusion ×1 Adaptive Verification ×1 AngelSpec ×1 Budget Thermostat ×1 Nucleus Filtering ×1 Relay Trajectory ×1 Prefix Failure ×1 Bootstrapped Correlation ×1 Set-level Scoring ×1 Cardinality-specific Interaction ×1 Semantic Identifiers ×1 Evolutionary Indexing ×1 Posterior-driven Editing ×1 Semantic-Behavioral Alignment ×1 Filesystem ×1 Organization ×1 Search Economy ×1 Tool Design ×1 Shelf Generation ×1 Offline Serving ×1 Sharpness-aware ×1 Salience Recovery ×1 Reward Guided Decoding ×1 layer localization ×1 model adaptation ×1 transformer learning ×1 LLM Re-ranking ×1 Food Delivery ×1 Repeat-Order ×1 Shapley Value ×1 Interactive Analytics ×1 Token-level Routing ×1 Memory Foundation Model ×1 Fast Weight Programmers ×1 Dense Memory Network ×1 RL4Rec ×1 ListwiseRanking ×1 PolicyOptimization ×1 StructureAware ×1 KendallTau ×1 CreditAssignment ×1 Discriminative ×1 LLM-native advertising ×1 sidecar pattern ×1 conditional rewriting ×1 controllable ad intensity ×1 NaiAD ×1 Live-Stream ×1 Semantic Description ×1 Visual States ×1 Cross-Task ×1 SkillEvolution ×1 DecoupledCredit ×1 Curriculum ×1 Multi-field Compression ×1 Context-aware Routing ×1 Lightweight Front-end ×1 Specification Elicitation ×1 Behavioral Oracle ×1 Two-Stage Framework ×1 Explorative Modeling ×1 End-to-End Generation ×1 Pretraining Axis ×1 Exploration Scaling ×1 Mode Coverage ×1 Hamming Topology ×1 Bit-decomposable Quantization ×1 Structural Mismatch ×1 two-clock learning ×1 temporal convolution ×1 cohort bias ×1 online advertising ×1 routing tokens ×1 episodic buffer ×1 parametric consolidation ×1 Variable-Length ×1 Token-Set Prediction ×1 Code-as-CoT ×1 Physical Consistency ×1 Dual-Engine ×1 Blender Simulation ×1 Heterogeneous GNN ×1 Random Walk ×1 LLM Profiling ×1 Sparse Recommendation ×1 Rating Prediction ×1 Semantic Integration ×1 Surface Routing ×1 Multi-Surface ×1 Knowledge Retrieval ×1 Midtraining ×1 Alignment Durability ×1 agentic recommendation ×1 attention allocation ×1 LLM-based simulation ×1 reputation mechanism ×1 market mechanism ×1 Lossy Verification ×1 Distribution Distortion ×1 Collaborative Verification ×1 Truncation-based Verification ×1 Social Search ×1 Relevance Learning ×1 claim-centered retrieval ×1 evidence graph ×1 faceted taxonomy ×1 Logit Interpolation ×1 Return-to-go Credit Assignment ×1 Agentic Visual Reasoning ×1 Adaptive Reward ×1 Bayesian active learning ×1 coreset selection ×1 sequential design ×1 benchmark estimation ×1 Edge LLM Agents ×1 Calibrated Deferral ×1 Lightweight Convergence Probe ×1 File-System Agent ×1 Expert Subspace ×1 Geometric Complementarity ×1 Functional Redundancy ×1 role-playing agent ×1 personalized rubric ×1 multi-turn conversation ×1 diffusion transformers ×1 Future Interactions ×1 Memory Pretraining ×1 kNN Distribution ×1 Zero-shot Decision Making ×1 Algorithmic Trading ×1 Non-Autoregressive ×1 Prefix-Anchored Credit ×1 Training Environments ×1 Stateful Applications ×1 Cost-Aware Stopping ×1 Tool Acquisition ×1 Decision-Focused Learning ×1 Prefix Stopping ×1 Regret-Weighted Classification ×1 Evolving Contexts ×1 Reverse KL Decomposition ×1 Execution-Grounded RL ×1 MLE-Bench ×1 differential activation ×1 GLU-MLP ×1 neural pruning ×1 Adaptive Rounding ×1 Two-Sided ×1 Kronecker Product ×1 Cubic Time ×1 ID-based embedding ×1 industrial scale ×1 multi-turn RAG ×1 trace-based memory ×1 conversational QA ×1 Multi-Layer Fusion ×1 Instance-Aware ×1 Dimensional Collapse ×1 Semantic Enhancement ×1 serving optimization ×1 memory system ×1 Chunked RoPE ×1 context-window encoding ×1 Memory Reconstruction ×1 OOD Robustness ×1 Thinking Traces ×1 Cross-domain History ×1 Structured Evidence Ledger ×1 Verification-and-Repair ×1 Visual Question Answering ×1 generative IE ×1 product attribute extraction ×1 PLM correction ×1 LLM validation ×1 privacy-preserving ×1 Hierarchical Reranking ×1 Query Optimization ×1 Financial NLP ×1 action menu ×1 graph retrieval ×1 graph-retrieval ×1 multi-hop ×1 knowledge-alignment ×1 contrastive-learning ×1 Kalman Filter ×1 Uncertainty Calibration ×1 In-model indexing ×1 Neural scoring ×1 Hierarchical index ×1 Index balancing ×1 Large-scale retrieval ×1 End-to-end training ×1 Physical World Model ×1 Zero-Shot Transfer ×1 Pair-Space Generation ×1 Latent Regularization ×1 Quantile-Quantile Matching ×1 Gaussianity ×1 Heavy-tail Control ×1 Action Batching ×1 Proactive Service ×1 Multi-Reference ×1 Hierarchical Coverage ×1 Sparse Token Selection ×1 Reliability Memory ×1 LLM Steering ×1 Weyl's Inequality ×1 TTRL ×1 rubric optimization ×1 Memory-augmented Agents ×1 User Understanding ×1 Personality Traits ×1 ShadowDancer ×1 Action Representation ×1 Cross-Shadow Prediction ×1 Dynamics Control ×1 knowledge internalization ×1 Temporal Heterogeneous Graphs ×1 Graph Transformers ×1 Rotary Temporal Attention ×1 Dual-Branch Architecture ×1 Type-Conditioned Gating ×1 Relative Time Encoding ×1 Query Intent Generation ×1 Discovery-Augmented Search ×1 dense-retrieval ×1 translate-train ×1 open-data ×1 hard-negative-mining ×1 multi-view reasoning ×1 Modality Imputation ×1 FailureTaxonomy ×1 InteractionModel ×1 FaultLocalization ×1 streaming evaluation ×1 Failure Modes ×1 Compute Tradeoffs ×1 Multi-View Attention ×1 Cross-Field Interaction ×1 Subspace Token Mixing ×1 Hierarchical Compression ×1 Persona Simulation ×1 Delphi ×1 Long-Video Understanding ×1 Latent Evidence ×1 Cross-Frame Integration ×1 Relational Distillation ×1 Feedback-driven ×1 belief simulation ×1 persona ×1 DFOT ×1 Privileged Distillation ×1 LLM Over-Trust ×1 Population Search ×1 Preserve-and-Extend ×1 Natural Selection ×1 self-refine ×1 reflexion ×1 repeated-sampling ×1 reasoning-evaluation ×1 llm-inference ×1 RL+distillation ×1 Dynamic Graphs ×1 Streaming Data ×1 Macro-Micro Graph ×1 Query-driven Retrieval ×1 Dynamic Programming Decoding ×1 Life Sciences ×1 schema-guided extraction ×1 enterprise documents ×1 agentic extraction ×1 cost evaluation ×1 Fine-Grained ×1 Image Retrieval ×1 CIR ×1 Two-Stage Finetuning ×1 Dataset Pipeline ×1 Heterogeneous Ranking ×1 Model Observability ×1 Cross-Segment AUC ×1 supply chain optimization ×1 end-to-end decision ×1 policy coupling ×1 shared representation ×1 replenishment planning ×1 UNICON ×1 graph neural network ×1 cross-disciplinary generalization ×1 numerical intelligence ×1 Policy Tree ×1 Anticipatory Execution ×1 Decoding Latency ×1 Transient Events ×1 Style Bias ×1 Cross-Teacher ×1 Multi-Speaker Dialog ×1 Turn Action ×1 Stage Replay ×1 Causality ×1 Divergence ×1 Salience Bias ×1 Commonsense Reasoning ×1 Knowledge Suppression ×1 Generative Disentanglement ×1 Local-Life Service ×1 Industrial System ×1 Co-engagement ×1 topology adaptation ×1 inference-time repair ×1 Long-Term Planning ×1 off-policy learning ×1 large action spaces ×1 mixed effects ×1 policy-weighted log-likelihood ×1 Gradient-free ×1 Low-rank ×1 On-device ×1 Attention Momentum ×1 Late-Blooming Tokens ×1 Dynamic Rectification ×1 Automated Metrics ×1 Quantized LLMs ×1 Memory Editing ×1 Scenario Annotation ×1 Enterprise RAG ×1 Data Asset Retrieval ×1 Hierarchical Representation ×1 GLM ×1 RCA ×1 Benchmark Misalignment ×1 PR-Issue Alignment ×1 Multi-Reference Editing ×1 Visual Consistency ×1 reader highlights ×1 length bias ×1 Request-Oriented Computing ×1 Deep Cross Attention ×1 Generalized Layer Masking ×1 text conditioning ×1 structured prompts ×1 visual generation ×1 Inference-Time Reranking ×1 Hierarchical Poisson Factorization ×1 Natural Language Bridge ×1 node classification ×1 link prediction ×1 homophily ×1 cross-task ×1 Multimodal Distillation ×1 Visual Attribution ×1 On-Policy Learning ×1 Target Reconstruction ×1 visual sampling ×1 query-grounded ×1 long video understanding ×1 advantage fusion ×1 dynamic pruning ×1 token-level sparsity ×1 kernel co-design ×1 width pruning ×1 exploratory search ×1 graph refinement ×1 dynamic index ×1 On-Policy Optimization ×1 Support Reshaping ×1 Concept Activation Vectors ×1 Sparse Autoencoders ×1 Image Similarity ×1 Example-based Explanation ×1 Adaptive Compute ×1 analytic memory ×1 agent systems ×1 attribute-value extraction ×1 Validation ×1 intent prediction ×1 code spotlight ×1 Open-source Framework ×1 Futile Reasoning ×1 Capability Alignment ×1 Refusal Training ×1 Reasoning Primitives ×1 multimodal representation ×1 recommendation system ×1 food delivery ×1 adaptive fusion ×1 Preference-based RL ×1 Latent Variable ×1 Reasoning Trace ×1 Multi-modal MoE ×1 ReBA ×1 Auxiliary Loss ×1 Markov Hypergraph ×1 Prototype-based ×1 QASP ×1 Recall Prediction ×1 Hypothetical Prompt Embeddings ×1 Question-Answer Gap ×1 Index-time Optimization ×1 negative sampling ×1 regional constraint ×1 batch sampling ×1 Bandit Routing ×1 residual attention ×1 Next POI ×1 Advantage Calibration ×1 Streaming Sequence Model ×1 Model-Infrastructure Co-design ×1 Production Scale ×1 KV Caching ×1 Advantage Asymmetry ×1 zero-token memory ×1 graph-based retrieval ×1 provenance-preserving ×1 evidence calibration ×1 memory utilization ×1 Know-Act gap ×1 Role-Break ×1 Task Sequencing ×1 skill memory ×1 tool-augmented ×1 Adaptive Rollout ×1 Plateau Detection ×1 Progress-Aware ×1 DENSEWORLD ×1 Factorized Prediction ×1 Global South ×1 Token-level Weights ×1 Persistent Consistency ×1 Latent Language Model ×1 Block-Causal ×1 tool selection ×1 deterministic gating ×1 executability ×1 Dual-Anchored ×1 Multi-source Supervision ×1 Vision-in-the-Loop ×1 Event Graph ×1 Skill Generation ×1 Rollback Reward ×1 Progressive Generation ×1 Task-State Management ×1 MEA Loop ×1 Multi-Role ×1 Guided Retrieval ×1 World Proxy ×1 Agent Improvement ×1 Interactive Agent ×1 Information Transition ×1 click preference ×1 Skill-Use Training ×1 Verified Data Generation ×1 Multi-Agent Data Synthesis ×1 Deep-Research Agent ×1 Selective Fetching ×1 BQL ×1 Tevatron ×1 Megatron ×1 Decoder-only ×1 Unified Embedding ×1 enterprise QA ×1 residual error sampling ×1 lossless compression ×1 tensor compression ×1 checkpoint compression ×1 Target-Aware Retrieval ×1 AgentOps ×1 Streaming Recommendation ×1 Pretrain-Transfer ×1 Knowledge Decoupling ×1 label-free quality signal ×1 region grafting ×1 coupling guard ×1 S-RCT ×1 Within-Subject Design ×1 Error Decomposition ×1 floating-point underflow ×1 structured memory ×1 edge inference ×1 state injection ×1 O(1) prefill ×1 context importance ×1 material difference ×1 prefix compression ×1 Spurious Signals ×1 Input-Groundedness ×1 Token-level Filtering ×1 LLM-VLM Training ×1 Deterministic Evaluation ×1 Semantic Sufficiency ×1 Cross-Conditional Reconstruction ×1 Symbolic Verification ×1 Best-of-K ×1 Rule Hybridization ×1 checkpoint ×1 exactly-once ×1 TLA+ ×1 LangGraph ×1 CrewAI ×1 hindsight critique ×1 search-augmented ×1 multi-turn search ×1 Missing Modality ×1 Point-of-Interest ×1 tool invocation ×1 belief state ×1 intent fluctuation ×1 Experience Evolution ×1 Simulated Users ×1 Digital Products ×1 Population-scale ×1 Relevance Evaluation ×1 Stratified Sampling ×1 LiveMem ×1 GDN2 ×1 memory state continuity ×1 context turnover ×1 Offline Distillation ×1 execution memory ×1 LLM-agnostic ×1 AgentIF ×1 diagnostic ×1 Tool-Augmented Dialogue ×1 Longitudinal Reasoning ×1 Screenshot-Only ×1 Iterative Training ×1 Reduced-Order State ×1 Memory-Reward Trap ×1 Utility Learning ×1 Adaptive Validation ×1 Drift-Aware Consolidation ×1 Agent Backdoor ×1 Prefix Pruning ×1 Hierarchy-Aware Distillation ×1 Tool-Integrated Reasoning ×1 Hindsight Self-Distillation ×1 Turn-Level Credit Assignment ×1 Unpaired Learning ×1 Modality-Agnostic ×1 Document Interface ×1 Compositional Query ×1 Requirement Alignment ×1 R-GRPO ×1 E-Commerce Reranking ×1 Near Miss ×1 Retrieval Gap ×1 Utilization Gap ×1 Evidence-Driven Stopping ×1 Trajectory Diagnosis ×1 field-aware ×1 MLP scoring ×1 lifelong agents ×1 multi-field ×1 Ordinal Regression ×1 Computational Efficiency ×1 Bellman RTG ×1 feedback loops ×1 Domain Generalization ×1 Zero-Shot Recommendation ×1 Gromov-Wasserstein ×1 Adversarial Alignment ×1 Multi-source Recommendation ×1 Trust Gate ×1 State-Conditioned Readout ×1 Over-inference ×1 Self-monitoring ×1 Greek NLP ×1 Dense retrieval ×1 open-weight ×1 LLM Prior ×1 Comment Recommendation ×1 Demographic Segmentation ×1 prefiller-decoder ×1 parallel training ×1 KV injection ×1 Skill Entropy ×1 Skill-Switching ×1 Recursive Synthesis ×1 Verified Data ×1 OOD Recommendation ×1 Latent Environment ×1 Identifiable Representation ×1 Causal Representation ×1 Golden Negative Trajectories ×1 Reflective Reasoning ×1 Policy Transition ×1 domain mismatch ×1 hybrid search ×1 scientific QA ×1 Long-Horizon Dynamics ×1 CycleBench ×1 Long-Sequence Recommendation ×1 Target-Aware Modeling ×1 Interest Compression ×1 Balanced Parallel Quantization ×1 3D Scene Understanding ×1 Semantic-Aware ×1 Sparse Probing ×1 Outcome Calibration ×1 Video-DR ×1 Decoupled Perception-Exploration ×1 VIDEODR-BENCH ×1 Preference Consistency ×1 Kemeny Optimization ×1 Coarse-to-fine ×1 Editable Assets ×1 Hierarchical Representations ×1 metadata traces ×1 vision encoders ×1 OOD generalization ×1 generated image detection ×1 Referential Dangling ×1 Context Pruning ×1 Dependency Preservation ×1 Attack ×1 Strategy Library ×1 State-Matched Distillation ×1 Multi-Turn Agents ×1 Privileged Guidance ×1 multimodal fusion ×1 reliability diagnosis ×1 heteroscedastic uncertainty ×1 modality restoration ×1 inverse-variance fusion ×1 intent recognition ×1 Step Reward ×1 Answer Backtracking ×1 Bayesian Belief Update ×1 Turn-Level ×1 Agentic Runtime ×1 Verification-Gated Evolution ×1 Procedural Generation ×1 Reasoning Training ×1 Completion Supervision ×1 System-level Adaptation ×1 Inference-time Training ×1 Memory-augmented ×1 latent reconstruction ×1 process alignment ×1 Long-Form Video Description ×1 Embedding Ensemble ×1 Trajectory Evidence ×1 Verifiable Records ×1 Industrial AI ×1 economic-world-model ×1 multi-agent-simulation ×1 institutional-evolution ×1 agentic-economy ×1 Dual-Mode Embedder ×1 Dual Exploration ×1 Exploratory Reward ×1 Multi-Head Decoding ×1 Adversarial Calibration ×1 Learnability ×1 Multimodal Pretraining ×1 Knowledge Flow ×1 Modality Synergy ×1 Early Unification ×1 environment distribution ×1 difficulty ×1 World Rehearsal ×1 collaborative-harness-evolution ×1 program-aggregation ×1 agent-self-improvement ×1 scope-typed ×1 evidence-guided ×1 distributed ×1 two-sided platforms ×1 marketplace interference ×1 extreme-value theory ×1 experimentation ×1 global spatial awareness ×1 VLM benchmark ×1 embodied agents ×1 Temporal Decay ×1 What-Where-When ×1 ScrubJay-MEM ×1 Deterministic Compilation ×1 Screen Capture ×1 Routine Replay ×1 Cost Instrument ×1 Capability Pages ×1 Exploration Bonuses ×1 Neural Memory ×1 Reward Structure ×1 Partial Observability ×1 filtered vector search ×1 lakehouse ×1 Apache Iceberg ×1 file pruning ×1 distributed index build ×1 Rollout Intervention ×1 Hallucination Correction ×1 Multi-source Verification ×1 Consensus Scoring ×1 Multi-LLM Jury ×1 Video ×1 Inference-time Architecture ×1 Curvature Preconditioning ×1 Diagonal Preconditioning ×1 Stochastic Optimization ×1 cost-performance trade-off ×1 QPC ×1 Delta Distillation ×1 Path-level Localization ×1 Coordinated Rewriting ×1 Long-term Reasoning ×1 code-context ×1 data-engineering ×1 agentic-tasks ×1 cross-repository ×1 Layout Grounding ×1 Prefix-conditioned Factorization ×1 Multilingual Embedding ×1 Task-Conditional ×1 Teacher-Guided ×1 Embedding Adaptation ×1 multi-task ×1 Parallel-RL ×1 skill use ×1 progressive disclosure ×1 Graph Compression ×1 Contract Preservation ×1 streaming video understanding ×1 long-horizon memory ×1 two-tier architecture ×1 proactive interaction ×1 inherent interpretability ×1 concept bottleneck ×1 diffusion LM ×1 Steerling-8B ×1 online experimentation ×1 video recommendation ×1 user behavior logging ×1 popularity calibration ×1 item familiarity ×1 live-streaming ×1 delayed signals ×1 user segmentation ×1 multi-objective ranking ×1 modality ×1 citation grounding ×1 Mobile GUI Agents ×1 Delta Code ×1 HTML Generation ×1 SFT Data Construction ×1 Test-Time RL ×1 Confidence Sequences ×1 AIVAT ×1 Anytime-Valid Stopping ×1 Kronecker-factored Hessian ×1 adaptive rounding ×1 Embedding-free retrieval ×1 Structured documents ×1 Model Context Protocol ×1 Conversational Commerce ×1 Hybrid Architecture ×1 Transparent Ranking ×1 hybrid ranker ×1 topology bound ×1 Token-Level Supervision ×1 Adaptive Gating ×1 Event Counting ×1 Diagnostic Benchmark ×1 policy coverage ×1 Multimodal Trading ×1 Financial Fusion ×1 Noise-Robust Regularization ×1 Modality-Aware Fusion ×1 personalization audit ×1 modality weighting ×1 per-user weighting ×1 Emotion Control ×1 Decoupled Field ×1 late-interaction retrieval ×1 exclusion query ×1 query-side module ×1 exclusion inversion ×1 benchmark contamination ×1 decoding intervention ×1 SA-PPG ×1 RailCap ×1 VoiceAgent ×1 FormFilling ×1 RuleBased ×1 EndToEndEval ×1 Hindi ×1 programmatic tool calling ×1 JSON tool calling ×1 BFCL v4 ×1 parallel fan-out ×1 Rollout Distillation ×1 Generate-and-Rank ×1 Unified Cascade ×1 Document Review ×1 Rule-Intensive ×1 GB-T Standards ×1 Counterexample Generation ×1 Deepfake Detection ×1 Forensic Reasoning ×1 Nested Attributes ×1 Personalized Query Refinement ×1 Intent Elicitation Graph ×1 Clarification Policy ×1 unlearnable examples ×1 copyright protection ×1 adversarial perturbations ×1 Zero-Shot Control ×1 Symmetry Consensus ×1 observation modes ×1 cost-quality tradeoff ×1 representation selection ×1 communication pruning ×1 Computation Budget ×1 comparative evolution ×1 reaction-norm mutation ×1 cross-lineage hybridization ×1 archive-based search ×1 Shadow Removal ×1 Physics-Informed ×1 Candidate Selection ×1 pre-inference routing ×1 document extraction ×1 difficulty prediction ×1 token-level steering ×1 co-writing ×1 TTT ×1 multimodal search ×1 vector indexing ×1 incremental encoding ×1 Explainable QA ×1 Knowledge Gap Detection ×1 visual tool-use ×1 causal audit ×1 multimodal LLMs ×1 LLM-driven search ×1 TPDSL ×1 compute-optimal training ×1 interaction exponent ×1 sparse grid ×1 Similarity Calibration ×1 Score Mapping ×1 Synthetic Query Probing ×1 Reward Construction ×1 Ranking-based Reward ×1 MIST ×1 SC2W ×1 Context Robustness ×1 Stateful Agent ×1 Cross-Device Interaction ×1 Engagement Evidence ×1 Compact State ×1 Error Lifecycle ×1 Critical Error Detection ×1 Multi-granularity Compression ×1 retrieval augmented generation ×1 reference tokens ×1 state-of-the-art ×1 Reasoning Pivot ×1 Cross-lingual Transfer ×1 conversational commerce ×1 domain knowledge ×1 explanation ×1 Evidence-Linked ×1 Healthcare AI ×1 Visual Composition ×1 Object-centric Learning ×1 Graph Attention ×1 Self-supervised Fine-tuning ×1 Cross-domain Generalization ×1 Post-Training Adaptation ×1 Model Governance ×1 Causal Memory ×1 Dual-Polarity ×1 Failure Repair ×1 YOLO ×1 Object Detection ×1 Adapter Placement ×1 Constraint Planning ×1 program evolution ×1 ML workflow optimization ×1 reasoning-driven search ×1 Syntax ×1 Positional Embeddings ×1 Dependency Parsing ×1 GLUE ×1 Architecture-Aware ×1 False Discovery Control ×1 Agent Task Protocol ×1 Multidimensional Metrics ×1 Execution Trace ×1 venue recommendation ×1 visibility bias ×1 market census ×1 end-to-end training ×1 long-horizon prediction ×1 endpoint objective ×1 non-recursive architecture ×1 compute routing ×1 nugget extraction ×1 long-context RAG ×1 latency-efficiency ×1 position alignment ×1 Creativity ×1 Graph Edit Distance ×1 Agentic Patterns ×1 Entity Reconciliation ×1 Skill Taxonomy ×1 Dynamic Retrieval ×1 Geo-spatial ×1 Ultra-long sequence ×1 Token merge ×1 Full attention ×1 Ad recommendation ×1 Neuron Pruning ×1 Transfer Attack ×1 Interactive Narrative ×1 Long-Horizon Consistency ×1 Hybrid Optimization ×1 CMA-ES ×1 Structure–Parameter Decoupling ×1 Routing Consistency ×1 Model Internal States ×1 Neuron Plasticity ×1 Language Preservation ×1 Multimodal Instruction Tuning ×1 Parameter-Efficient Tuning ×1 prefetching ×1 reviewed commits ×1 Automated Algorithm Design ×1 Code Evolution ×1 Genetic Programming ×1 Executable Primitives ×1 Proximal Gradient ×1 Closed-loop Diagnosis ×1 Knowledge Consolidation ×1 Memory Pollution ×1 Lifecycle Revocation ×1 Staleness ×1 Conflict Keys ×1 TEPA ×1 Object Hallucination ×1 Visual Attention ×1 Detect-Mitigate ×1 Multimodal AI ×1 Telecom Analytics ×1 Progressive Fine-Tuning ×1 Hierarchical Control ×1 Intent Contracts ×1 Persona Policy ×1 Trajectory Auditing ×1 Live-Web Evaluation ×1 Adaptive Fusion ×1 Exact Top-K ×1 Rhetorical Chunking ×1 Legal QA ×1 Audio Embedding ×1 Empirical Evaluation ×1 Meta-Evolution ×1 BALROG ×1 Task-Specific ×1 Gender Bias ×1 NMT ×1 Romanian ×1 Push Recommendation ×1 business simulation ×1 diagnostic metrics ×1 latent feedback ×1 data-efficient pretraining ×1 multi-pass training ×1 recurrence ×1 expert merging ×1 acceleration ×1 Unified Representation ×1 Streaming Summarization ×1 Dialogue Summarization ×1 Missing Evidence ×1 Marginal Value Pruning ×1 Stage-aware Pruning ×1 Heuristic Pruning ×1 Structure Preservation ×1 Collaborative Embedding ×1 LLM robustness ×1 stress testing ×1 logit masking ×1 circumlocution ×1 instruction alignment ×1 MoL ×1 Agent Communication ×1 Perspective Narrowing ×1 Grouped Differential Attention ×1 Multi-teacher Distillation ×1 Low-precision Training ×1 Persistent Search Drift ×1 Search-State Brief ×1 Search Control ×1 Code Refactoring ×1 rhetorical sensitivity ×1 scoring bias ×1 UserSimulation ×1 IntentControl ×1 DialogueSystem ×1 Fréchet Distance ×1 Feature Whitening ×1 ARC-AGI-1 ×1 Adaptive Capacity Allocation ×1 Recommendation Efficiency ×1 Dirichlet augmentation ×1 Rao-Blackwellization ×1 categorical generation ×1 Social Recommendation ×1 LLM Policy ×1 Path Exploration ×1 AgenticRecSys ×1 IntentEngine ×1 MetaControl ×1 RewardDualLoop ×1 TrafficFunnel ×1 IndustrialRecSys ×1 Real-OS ×1 Dropout ×1 Modality Missing ×1 Parameter-space exploration ×1 Variational learning ×1 financial agents ×1 runtime governance ×1 authority control ×1 structured contracts ×1 receipts ×1 reference-free reward ×1 checkpoint interpolation ×1 multilingual MT ×1 listwise LTR ×1 medical IR ×1 Test-Time Augmentation ×1 Semantic Rephrasing ×1 Hierarchical Query ×1 Task Interaction ×1 Node Ranking ×1 Subgraph Neural Networks ×1 calibration-free ×1 Token Credit ×1 spectral CF ×1 sign-aware ×1 negative feedback ×1 thought-level beam search ×1 hidden-state scorer ×1 Language-level RL ×1 Small LVLM ×1 Policy Iteration ×1 Cost-Effective Agent ×1 Skill Compression ×1 Minimum Description Length ×1 Structured Contracts ×1 Zip-on-Write ×1 Skill Harness ×1 Train-free ×1 Proxy Reward ×1 Proactivity ×1 Everyday-life Assistance ×1 Verbal Reinforcement Learning ×1 Preference Adaptation ×1 Textual Optimization ×1 Task-Specific Compression ×1 LLM Strategy Generation ×1 Rényi divergence ×1 Self-referenced target ×1 Combo Shopping ×1 Budget Constraint ×1 Coupon Optimization ×1 Basket Construction ×1 Agentic Auto-Encoder ×1 Video Knowledge Graph ×1 Video Reconstruction ×1 Recursive Delegation ×1 Autonomous Coding ×1 Joint Search-Recommendation ×1 Query Semantics ×1 PLE ×1 Gram Newton-Schulz ×1 AI Coding Agent ×1 Specification-first ×1 Refactoring ×1 Verification Loop ×1 Case Study ×1 Agentic Software Engineering ×1 Anchor-Based ×1 Pointwise ×1 Contrastive Scoring ×1 Reasoning Compression ×1 Slow-Thinking Recommender ×1 Fisher-weighted Merging ×1 LLM Acquisition ×1 Reward-SNR ×1 Offline Policy Learning ×1 Structured Hypothesis Embeddings ×1 Gated Fusion ×1 Visual Utility ×1 Unstructured KE ×1 Hybrid Rollout ×1 Composability ×1 Business Ideation ×1 MLLM-as-Judge ×1 LLM caching ×1 semantic query cache ×1 session management ×1 Redis ×1 embedding dedup ×1 mixed rollout ×1 on-device agent ×1 Behavioral Control ×1 Activation Directions ×1 Psychometrics ×1 Prompt Registers ×1 B-data ×1 Group Recommendation ×1 Tie-Breaking ×1 HR@K ×1 NDCG@K ×1 cohort batching ×1 device-resident control ×1 agent runtime ×1 Teacher Distillation ×1 Unified Ranking ×1 Self-Refutation Loop ×1 Citation Validity ×1 Figure Editability ×1 Deterministic Checks ×1 Generative Data Augmentation ×1 Tail Augmentation ×1 Tabular VAE ×1 Operational Validity ×1 Imbalanced Regression ×1 Time-aware Recommendation ×1 Graph Reconstruction ×1 Strong-to-Weak ×1 Theory-of-Mind ×1 post-calibration ×1 reliability reranking ×1 label-wise monotone projection ×1 fallback routing ×1 relevance prediction ×1 Multilingual Agents ×1 Policy Retention ×1 Cross-Lingual Evaluation ×1 Action Trace ×1 Long-Horizon Generation ×1 Linear Scaling ×1 Agent Interaction ×1 Reward Fairness ×1 Meta-Harness Optimization ×1 Poster Generation ×1 Reasoning behaviors ×1 Behavioral Lift ×1 Process reward ×1 Confidence calibration ×1 Similarity Bias ×1 Logical Reasoning ×1 Integer Linear Programming ×1 MCQ ×1 Reward Reachability ×1 Hint-Based RL ×1 Simulator Collapse ×1 LLM User Simulator ×1 Verbalized Sampling ×1 Policy Entropy ×1 Schema Exploration ×1 Rule-based Correction ×1 Multi-path SQL Generation ×1 LLM Prompting ×1 Top-k Allocation ×1 Effective Sample Size ×1 Optimizer's Curse ×1 Propensity Estimation ×1 Pre-Attention Spikes ×1 Inter-Spike Plateaus ×1 Output Gating ×1 LLM Interpretability ×1 Memory Decoder ×1 ranking reward distillation ×1 Transfer Reliability ×1 Procedure Memory ×1 8-bit Optimizer ×1 Layer Streaming ×1 Answer Selection ×1 zero-inflation ×1 multi-tier coupons ×1 Semantic Segmentation ×1 Structured Retrieval ×1 schema evolution ×1 Gradient Manipulation ×1 Matrix Geometry ×1 Orthonormal Updates ×1 lexical diversity ×1 world models ×1 agent players ×1 latent context ×1 LLM Serving Cost ×1 Conversational Agents ×1 Graph-Aware LLM ×1 Attention Bias ×1 Multimodal Summarization ×1 VLM Routing ×1 Open-Set ×1 Execution-Oriented ×1 CRM+RCCR ×1 Autoregressive Video Generation ×1 DMD ×1 Few-step Distillation ×1 Resource Amplification ×1 Cost Safety ×1 Attribute Factorization ×1 Fashion-CLIP ×1 Controllable Retrieval ×1 Text-Image Gating ×1 Multi-turn Simulation ×1 Skill Governance ×1 Multi-hop reasoning ×1 API retrieval ×1 parametric knowledge ×1 context addiction ×1 gradient allocation ×1 selective question answering ×1 conformal risk control ×1 domain-specific ×1 clinical LLM ×1 Agentic Transactions ×1 ACID ×1 conversational advertising ×1 judgment model ×1 tool coevolution ×1 Intervention Response ×1 Protocol Audit ×1 Behavioral Policy ×1 Selective Response ×1 Frontier Models ×1 AI R&D ×1 Heterogeneous Graph Transformer ×1 Task relevance ×1 Selective distillation ×1 Token-level supervision ×1 Targeted Regularization ×1 Selection Bias ×1 Autoregressive Tree ×1 Lossless Speedup ×1 HRM ×1 1B LLM ×1 permissible data ×1 Danish ×1 Idle Window ×1 Robust Learning ×1 Ensemble Ranking ×1 Prediction Noise ×1 Rank Consistency ×1 session-based recommendation ×1 time-aware MLP ×1 stopping policy ×1 Search-R1 ×1 structured judge ×1 retrieval reduction ×1 concept drift ×1 OCSVM ×1 adaptive retraining ×1 malware classification ×1 minibatch k-means ×1 boosting ×1 adversarial sequences ×1 weak learning ×1 Token Frequency Bias ×1 Generative Embedding ×1 Synthetic Persona Pretraining ×1 Persona Binding ×1 Constitution Following ×1 Humanoid Motion Tracking ×1 Temporal Transformer ×1 Bangla ×1 Knowledge-Reasoning Decoupling ×1 Edge Computing ×1 Virtual Agents ×1 Cognitive Architecture ×1 Matrix Multiplication ×1 Input-Adaptive Pruning ×1 Attention Redundancy ×1 quantisation ×1 Over-refusal ×1 Intent-Group ×1 Longitudinal Conversations ×1 Proactive AI ×1 Pretraining Corpus ×1 Knowledge Boundaries ×1 TPP ×1 Gricean Maxims ×1 Referent Specificity ×1 Clinical AI ×1 Decomposer ×1 Multilingual RAG ×1 Deferred Translation ×1 Graph Alignment ×1 PCVR ×1 feature tokenization ×1 target-aware ×1 HyFormer ×1 MLLM Router ×1 Ensemble Learning ×1 Image Classification ×1 Heterogeneous Models ×1 Cross-Dataset ×1 Bash quoting ×1 execution transport ×1 command generation ×1 final-state validation ×1 Poisoning Detection ×1 task vector ×1 multimodal ICL ×1 activation intervention ×1 Narrative Reasoning ×1 Cultural Understanding ×1 Japanese ×1 Associative Recollection ×1 Verbalization ×1 Feature Explanation ×1 Part-aware ×1 Structured Latent ×1 Subtractive Composition ×1 Time-Series ×1 Financial Forecasting ×1 CalibRank ×1 Scientific Figures ×1 Behavioral Reliability ×1 Hybrid Reranking ×1 Expert Finding ×1 Authority Modeling ×1 Multi-View Summarization ×1 Quotation Faithfulness ×1 Streaming VLM ×1 Fast Weights ×1 Real-time Perception ×1 Tabular-to-image ×1 Self-Organizing Maps ×1 Feature Encoding ×1 sample selection ×1 outlier detection ×1 robust learning ×1 Query Engine ×1 Schema Annotation ×1 SQL Extension ×1 Document Analytics ×1 ANN Search ×1 Latency ×1 Paper Replication ×1 Rubric Judge ×1 Mixture of Training ×1 Scaffolded Pretraining ×1 Modular Training ×1 Critical Path ×1 training data influence ×1 gradient trajectory ×1 Pythia ×1 PolyPythia ×1 Adaptive Regularization ×1 Dual Ascent ×1 Multi-hop retrieval ×1 Query reformulation ×1 Evidence-guided retrieval ×1 evidence synthesis ×1 task routing ×1 multi-source synthesis ×1 runtime recovery ×1 checkpointing ×1 Hierarchy-Aware ×1 Image-Text Alignment ×1 Document Grounding ×1 dLLMs ×1 Consistency Distillation ×1 Confidence Adaptive KL ×1 Self-rollout ×1 creative writing ×1 genre expansion ×1 Trajectory Value ×1 Message Filtering ×1 DHD ×1 Wrong-but-Useful ×1 small-batch decoding ×1 expert loading ×1 parameter sharing ×1 evidence aggregation ×1 Action-Conditioned ×1 On-Policy Matching ×1 Importance Weighting ×1 Distribution Sharpening ×1 Knowledge Extraction ×1 Ultra-long Documents ×1 Memory Substrates ×1 retrieval diversity ×1 FictionalQA ×1 generation correctness ×1 Cognitive Risks ×1 Human Agency ×1 Controllability ×1 Long-horizon agent ×1 root-cause localization ×1 Constraint Enforcement ×1 transaction embeddings ×1 zero-shot prediction ×1 sequence encoder ×1 adaptive pruning ×1 batch inference ×1 reasoning model ×1 activation memory ×1 Cloud-Device Collaboration ×1 Ad Recommendation ×1 DPP Diversity ×1 Lightweight Ranking ×1 Loss-aware Reconstruction ×1 Low-bit Inference ×1 Proactive Exploration ×1 Tree-structured Context ×1 Hierarchical Graph ×1 Spreadsheet Reasoning ×1 State Machine ×1 Model-Based Planning ×1 ARC-AGI-3 ×1 3D World Building ×1 MCP Tools ×1 text-to-SQL ×1 deterministic execution ×1 sufficiency detection ×1 single forward pass ×1 transform coding ×1 reverse water-filling ×1 Open-ended Search ×1 Retrieval Augmented ×1 Cross-Dataset Generalization ×1 zero-shot forecasting ×1 spectral periodicity ×1 dilated convolution ×1 quantile decoder ×1 tiny model ×1 Open-Weight ×1 Instruction Generation ×1 Continuous Environments ×1 Delegation Security ×1 Authorization ×1 Composition Closure ×1 Data-Centric AI ×1 Language-Specific Competency ×1 X-Lingual Reasoning ×1 Hybrid-Thinking ×1 Response Alignment ×1 Black-box RL ×1 Prefix Tree ×1 Training-Infrastructure ×1 Trajectory Graph ×1 Failure Detection ×1 Relational Propagation ×1 Agent Reliability ×1 Multimodal Ranking ×1 Survival Weighting ×1 Conversion Optimization ×1 Edge Serving ×1 Bandwidth-Adaptive ×1 CPU-GPU Offloading ×1 Agentic Workloads ×1 Information Isolation ×1 external memory ×1 reader adaptation ×1 knowledge reuse ×1 Graph-RAG ×1 MoE quantization ×1 Adaptive Parallelism ×1 Policy Alignment ×1 Operator Scheduling ×1 GPU Utilization ×1 Multi-Tenant Serving ×1 Simulation-guided ×1 PyTorch Backend ×1 Low-Rank Subspace ×1 Memory-Efficient ×1 Harness-Aware Training ×1 HSA ×1 Live E-commerce ×1 Qwen3.6-35B-A3B ×1 GRPO-GDPO ×1 staleness filtering ×1 multimodal PTR ×1 candidate pruning ×1 Google Discover ×1 advantage reweighting ×1 Looped LLM ×1 Compositional API ×1 Entropy Reduction ×1 Situation State ×1 Constraint Propagation ×1 Harnessed RL ×1 Prompt-Parameter Co-evolution ×1 GPU Memory Efficiency ×1 incremental SVD ×1 semantic substrate ×1 embedding drift ×1 orthogonal Procrustes ×1 latent semantic indexing ×1 Source-to-Sink ×1 DeepSeek Harness ×1 Multi-Objective Evaluation ×1 Operational Cost ×1 Grafting ×1 Source-Style Shift ×1 TF-IDF Routing ×1 Security Evaluation ×1 Harness-Native ×1 Multi-Graph Learning ×1 LLM-Augmented ×1 Video Distillation ×1 Temporal Confidence ×1 Branch Pruning ×1 Impression Share ×1 Counterfactual Prediction ×1 Delivery Capacity ×1 Regulation ×1 Hardness Calibration ×1 Semantic Grounding ×1 VLM Evaluation ×1 Outcome-Oriented ×1 control variates ×1 semi-supervised ×1 conditional performance ×1 cheap signals ×1 Pseudo-Query Generation ×1 Retriever Training ×1 E2E Workflows ×1 Market-Validated ×1 Attention Kernel ×1 INT8 Quantization ×1 Continual Recommendation ×1 LLM Enhancer ×1 Semantic Representation ×1 TREC ×1 Complementary Products ×1 UniDot ×1 CTR-CVR ×1 Factorization Machine ×1 Multi-path Mutual Learning ×1 unbiased learning ×1 learning to rank ×1 agentic recommender systems ×1 two-sided receptivity ×1 delegation asymmetry ×1 graded response model ×1 online dating ×1 Evidence-Based Review ×1 Scientific Workflows ×1 scalable oversight ×1 Complementary Links ×1 Layer-wise Exits ×1 Compact LLM ×1 Discriminative LLM-RS ×1 Reward Filtering ×1 Entity-Aligned Retrieval ×1 Semantic Distillation ×1 low-resource language ×1 behavioral metrics ×1 Self-rewarding ×1 Cross-agent Supervision ×1 Model Diversity ×1 Unsupervised Reasoning ×1 Executable Environment ×1 Verifier Grounding ×1 Behavior Analysis ×1 Residual Reconstruction ×1 Hybrid Agent ×1 Policy Invariance ×1 Potential-based ×1 difficulty estimation ×1 graph propagation ×1 online variational inference ×1 sample scheduling ×1 Chain-of-Experience ×1 test-time learning ×1 correspondence ×1 identifiability ×1 optimal matching ×1 two-sided validation ×1 Localized Supervision ×1 router entropy ×1 Neural memory ×1 Test-time learning ×1 Fast weights ×1 Efficient inference ×1 MediaFM ×1 IPS ×1 publisher traffic ×1 click-through rate ×1 user experience ×1 Project-Specific Knowledge ×1 Entity-Grounded Retrieval ×1 Issue Resolution ×1 Knowledge Work ×1 Workspace Management ×1 Document Editing ×1 Video-Audio Generation ×1 Preference Dataset ×1 Visually Rich Document Retrieval ×1 Evidence Path ×1 Document Embedding Preservation ×1 Straight-Through ×1 debiased inference ×1 measurement error ×1 LLM annotations ×1 CP decomposition ×1 semiparametric inference ×1 multiple proxies ×1 Verification-Gated ×1 PLC ×1 Runtime Evaluation ×1 Phase-wise Self-Bootstrapping ×1 CountSketch ×1 Listwise Planning ×1 memory-based agents ×1 variance ×1 task order ×1 underspecification ×1 Strategy-level capability ×1 Execution-level capability ×1 Empirical Analysis ×1 MRL embeddings ×1 Scalar Quantization ×1 recall optimization ×1 Human-Centered AI ×1 Interactive Information Access ×1 Trust ×1 Stateful Workflows ×1 Thought Block Chain ×1 Long-Context QA ×1 Citation Grounding ×1 Momentum ×1 Anisotropic ×1 EnvHarness ×1 Receding-Horizon Planning ×1 Task-Conditioned Architecture ×1 VLM planning ×1 evidence retrieval ×1 financial AI ×1 Multi-Intent ×1 out-of-fold ×1 Graph-adjacent Context ×1 HotpotQA ×1 Evidence Recovery ×1 Top-K Retrieval ×1 Multimodal VQA ×1 Evidence-Aware Calibration ×1 ScienceQA ×1 Educational AI ×1 CVE prioritization ×1 preference-adaptive fusion ×1 multi-view ×1 query-based ranking ×1 lightweight classifier ×1 headroom capture ×1 benchmark overfitting ×1 Audio Representation ×1 Autoregressive Prediction ×1 Next Patch Embedding ×1 Causal Transformer ×1 Spectrogram ×1 Cognitive Traps ×1 Inference-time Mitigation ×1 AdaptiveMem ×1 Aircraft Maintenance ×1 Synthetic Queries ×1 Multi-Scenario Ranking ×1 Long-context User Modeling ×1 SAIM ×1 Token Budget Balancing ×1 Open-Source Recipe ×1 Controlled Experiment ×1 Recommender Optimization ×1 Decision Tree ×1 Workflow Graph ×1 Runtime Verifier ×1 τ2-bench ×1 Agent Safeguard ×1 LLM recommendation ×1 editable memory ×1 retrieval-augmented ranking ×1 Dual-DAG ×1 Artifact ×1 Trace Accounting ×1 Hint-based Regret ×1 Personalized Search ×1 gradient boosting ×1 recourse ×1 leaf embedding ×1 Multi-view Learning ×1 Synthetic Task Prior ×1 Long-context reasoning ×1 Verifier reward ×1 business scenarios ×1 verifiable environments ×1 Training Algorithm ×1 Algorithmic Design ×1 Adversarial OCR ×1 LMM ×1 Subtask Decomposition ×1 Text Skills ×1 Code Skills ×1 Inference-Time Control ×1 NSMC ×1 Adapter Routing ×1 Conditional Alignment ×1 Utility-Preserving ×1 Task Model Induction ×1 Process Mining ×1 Trajectory Clustering ×1 Edge RAG ×1 Adaptive Systems ×1 CPU inference ×1 Evidence Arbitration ×1 Tool-augmented Decision ×1 Numerical Reasoning ×1 Planner-Renderer ×1 Instruction-based Image Editing ×1 decision tree ×1 split search ×1 Observational Data ×1 Hypothesis Prioritization ×1 Image Re-ranking ×1 Structured Evidence ×1 Listwise Verification ×1 Block-Sparse Attention ×1 Prefill Optimization ×1 FP8 Inference ×1 Long-Context Serving ×1 FormalTCS ×1 autoformalization ×1 theoretical computer science ×1 Lean theorem proving ×1 research automation ×1 null hypothesis ×1 statistical testing ×1 InsufficiencyBench ×1 Underspecified Queries ×1 corpus projection ×1 Interaction Effects ×1 Drug Combination ×1 Integrated Gradients ×1 Percentile Binning ×1 Clinical Prediction ×1 Manifold Drift ×1 ThermoDPO ×1 mobile inference ×1 S3D8 ×1 LFU ×1 quality-adjusted hit rate ×1 Spatial Planning ×1 Rule Following ×1 Disentangled Planning ×1 Agentic Trajectory ×1 Pandora's Box ×1 value of information ×1 cost-efficient inference ×1 auction-based allocation ×1 PEA-DPO ×1 Visual Insensitivity ×1 Physics-Informed Exploration ×1 Trend-Seasonal Decomposition ×1 Lightweight ×1 Interpretable ×1 AI clone ×1 backchannel ×1 nodding ×1 multimodal interaction ×1 co-presence ×1 listening behavior ×1 Software 3.0 ×1 LLM OS ×1 Agentic Computing ×1 Generalized Database ×1 Software Architecture ×1 Retrieval-Rerank ×1 Intent Memory ×1 cross-lingual fairness ×1 typological disparity ×1 Adaptive Evaluation ×1 Variance-Weighted Sampling ×1 Evaluation Efficiency ×1 post-hoc ×1 margin-controlled ×1 class imbalance ×1 music information retrieval ×1 E-commerce Live Streaming ×1 Persona Generation ×1 Dialogue Systems ×1 Delegation ×1 Interaction Effort ×1 Repository-level Benchmark ×1 Dynamic Instantiation ×1 Capacity Probes ×1 Pairwise Transition ×1 eSASRec ×1 H2O ×1 self-routing ×1 Item Embedding Refinement ×1 muP ×1 user acceptance ×1 survey dataset ×1 agent receptivity ×1 cross-cultural ×1 Predictive Factorization ×1 Orthogonality ×1 Query Clarification ×1 LLM prediction ×1 behavior modeling ×1 active data collection ×1 Knowledge Internalization ×1 Retrieval-Free QA ×1 Relation ×1 Self-Exchange ×1 FlashRelation ×1 Linear Relation ×1 Skill Utilization ×1 Codebook Update ×1 Exposure-aware ×1 Entity-Structure Index ×1 Text-to-Video ×1 Checklist ×1 Future Prediction ×1 MemGuide ×1 MemTools ×1 GUI grounding ×1 Memory Poisoning ×1 Provenance Ranking ×1 Content Screening ×1 LLM as Proposal ×1 Estimation of Distribution ×1 Medical AI ×1 explainable AI ×1 decision support ×1 faithfulness evaluation ×1 learning analytics ×1 Evaluation Agent ×1 Knowledge Poisoning ×1 Trust Index ×1 Quantization-Aware Healing ×1 MXFP4 ×1 Graph Engineering ×1 System Intelligence ×1 Task Organization ×1 Runtime State ×1 Uncertainty Preservation ×1 Calibration Data Selection ×1 DPQ ×1 semantic compilation ×1 incremental maintenance ×1 indexing ×1 Router Null Space ×1 Truthfulness ×1 LLM Recommenders ×1 Item Profiling ×1 Subjective Traits ×1 Analytic Learning ×1 Spectral Partitioning ×1 Exemplar-Free ×1 Class-Incremental Learning ×1 WY transform ×1 inference serving ×1 event-study ×1 endogenous time zero ×1 episode-selection bias ×1 negative controls ×1 web log analysis ×1 middle-tail ×1 holdback experiment ×1 engagement ×1 concentration ×1 residual quantization ×1 Document Authoring ×1 Structured Conditioning ×1 XML ×1 Behavior Denoising ×1 Structural Matching ×1 Sample Weighting ×1 Offline Calibration ×1 Critic-Generator Pipeline ×1 Hint-based RL ×1 Gaussian Guidance ×1 Rollout Efficiency ×1 Prompt-Tool-Middleware Patching ×1 EvoDAG ×1 Generalization-Aware Selection ×1 Audio-Visual Generation ×1 Cross-Shot Memory ×1 6-DoF Camera Trajectory ×1 Self-Gradient Forcing ×1 finite state machine ×1 next-step prediction ×1 workflow memory ×1 Graph Distillation ×1 Facet-aware Retrieval ×1 Coverage-aware Distillation ×1 Graph-RRF ×1 ERPO ×1 Query-KL ×1 Exploration-Stability ×1 Mathematical Discovery ×1 Shared Literature ×1 Long-Workflow ×1 Verifier-Grounded Experience ×1 Skill vs Gene ×1 specialist agents ×1 Sub-agent ×1 Skill Usage ×1 Answer Churn ×1 Backward Compatibility ×1 Index Updates ×1 no-CoT data ×1 Mixed SFT ×1 next-chunk reasoning ×1 reasoning post-training ×1 Query-Adaptive Width ×1 Orthogonal Adapter ×1 Ranking Stability ×1 Adversarial Evaluation ×1 Natural Integration ×1 User Satisfaction ×1 Safety Rules ×1 Knowledge Triage ×1 Long-Running Agents ×1 Embedding Indexing ×1 Dynamic scheduling ×1 Domain adaptation ×1 Validation-Gated Evolution ×1 Multi-Modal Retrieval ×1 Fashion QA ×1 Retrieval Reranking ×1 Retrieve-Rerank ×1 risk-aware reranking ×1 retrieval safety ×1 dual-head ranking ×1 tool risk annotation ×1 Scientific Workflow ×1 Workflow Completion ×1 World-Action Model ×1 Generative Control ×1 LASI ×1 Trigger-Induced Recommendation ×1 Pairwise Loss ×1 LLMRec ×1 Dual Experts ×1 interaction tax ×1 MoA ×1 verifier-scored optimization ×1 ensemble diversity ×1 model handoff ×1 trajectory interface ×1 cost-quality ×1 Skill Creation ×1 Procedure Admission ×1 Learning-to-Rank ×1 LambdaLoss ×1 Noise Rescheduling ×1 RQ-KMeans ×1 Attention Interpretability ×1 Axiomatic IR ×1 RankLLaMA ×1 Self-evolving memory ×1 Process constraints ×1 Typed memory banks ×1 Clinical agents ×1 Safety evaluation ×1 Request-gated tools ×1 Meta-Reasoning ×1 Recursive ×1 alignment tax ×1 RLM ×1 recursive subagents ×1 Continual Harness ×1 Critic Optimization ×1 DPPO ×1 Advantage Normalization ×1 UI-to-Code ×1 Visual Repair ×1 Submodular Scheduling ×1 Subword ×1 Attention Overload ×1 User Representation Learning ×1 Behavioral Tokenization ×1 Billion-scale ×1 Adaptive Quantization ×1 Frozen MLLM ×1 Step-Level Supervision ×1 Safety-Utility Balance ×1 Wasserstein Distance ×1 Opinion Summarization ×1 Scalable RL ×1 Data Regime ×1 Sample Weight Decay ×1 GPU Parallel ×1 Stabilizers ×1 Latent Intervention ×1 variance collapse ×1 distributional modeling ×1 Dual-Encoder ×1 Unified Search and Recommendation ×1 Full-Softmax ×1 skill graph ×1 counterfactual probing ×1 Bayesian calibration ×1 personalized PageRank ×1 Bayesian self-escalation ×1 Code Retrieval ×1 Prefix-Aware Decoding ×1 Rank Distillation ×1 E2EM ×1 Mine-Then-Train ×1 multilingual LLM ×1 skill evaluation ×1 text games ×1 cross-lingual consistency ×1 Verifiable Reasoning ×1 Unsupervised Post-Training ×1 product linking ×1 VLM distillation ×1 agentic VLM ×1 adaptive cascade ×1 Just-in-Time ×1 Emotion Recognition ×1 Error Augmentation ×1 Physics Reasoning ×1 Bilingual ×1 Query prediction ×1 Adaptive reasoning ×1 Model compression ×1 Audio-Visual Learning ×1 Task Disentanglement ×1 latency-accuracy tradeoff ×1 Expert Collaboration ×1 BI-RADS ×1 Breast Ultrasound ×1 Verifiable Skills ×1 harness evolution ×1 task stratification ×1 model transfer ×1 Generative Re-ranking ×1 Listwise Evaluation ×1 Survival Analysis ×1 Live-Streaming Advertising ×1 Temporal Adaptation ×1 Item Tokenizer ×1 Flow-based Model ×1 Codebook Guidance ×1 Visual Faithfulness ×1 zero-shot reasoning ×1 Voice Memory ×1 Dual-Brain Memory ×1 Affective Attribution ×1 Speech Agent ×1 Collapse ×1 Evidence Blindness ×1 Multi-view Atlas ×1 Game Engine ×1 Agentic Generation ×1 LLM-Enhanced CF ×1 Spectral Decoupling ×1 Non-principal Components ×1 Prediction-level Fusion ×1 Agentic Tool-Calling ×1 DAG Workflows ×1 Evaluation Reliability ×1 Diagnostic Framework ×1 Medical Flowcharts ×1 CRISP ×1 Token-Level Selection ×1 Vector Interpolation ×1 Unsupervised Calibration ×1 request-aware ×1 masking ×1 parameter-efficient ×1 model control ×1 frozen backbone ×1 Failure Mode ×1 Gradient Boosting ×1 Dynamic Ensemble ×1 Disentangled Learning ×1 data citation ×1 knowledge graphs ×1 training data attribution ×1 long-term user value ×1 causal effect ×1 actor-critic ×1 multi-objective fusion ×1 real-time video editing ×1 diffusion distillation ×1 appearance-motion decoupling ×1 character editing ×1 Concept Bottleneck Models ×1 answer selection ×1 rank AUC ×1 self-evolving LLM ×1 adversarial self-play ×1 Challenger-Solver-Judge ×1 zero data ×1 judge co-adaptation ×1 Agentic Editing ×1 LLM+VLM ×1 Long Video Editing ×1 MMLVE ×1 CPT ×1 Writing Trajectory ×1 Scientific Papers ×1 Pass@K ×1 Activation Map ×1 Reasoning Scaffold ×1 NLI Grounding ×1 rubric ×1 omni models ×1 Topology Reconstruction ×1 Probabilistic Alignment ×1 Stochastic Sampling ×1 supervisor-worker ×1 live-steering ×1 live-self-evolution ×1 Evidence-aware ×1 System Selection ×1 Answer Support ×1 Online Orchestration ×1 Progress Prediction ×1 sliding window attention ×1 Misinformation ×1 Reasoning Chains ×1 Triple Extraction ×1 Event Tokenization ×1 Asynchronous Caching ×1 Snapshot Resolution ×1 ACE Framework ×1 Agentic Data ×1 post-hoc sparsification ×1 Transformer Retrieval ×1 Feature Aggregation ×1 Model-based Retrieval ×1 Persistent Knowledge ×1 Cross-model Transfer ×1 Hamiltonian Dynamics ×1 Momentum Extrapolation ×1 State Space ×1 Autonomous Post-Training ×1 Experience Transfer ×1 Private Retrieval ×1 Deep Hashing ×1 Shortlist ×1 self-evolving AI ×1 evolution direction proposal ×1 industrial AI system ×1 Longitudinal Sensing ×1 Time-series Reasoning ×1 head importance ×1 transformer interpretability ×1 Deterministic Pipeline ×1 Tool Store ×1 Information Floor ×1 Model Gap ×1 meta-review ×1 UGC sparsity ×1 explainable recommendation ×1 Open World Object Detection ×1 Cross-Modal Calibration ×1 Multimodal Foundation Models ×1 OOD Detection ×1 Proactive Context Management ×1 Fine-grained RL ×1 friend recommendation ×1 multi-hash embeddings ×1 temporal neighbor sampling ×1 industrial recommender ×1 Vector Memory ×1 causal circuits ×1 Counterfactual Verification ×1 Recovery Language ×1 Behavior-Aware Verification ×1 Attributable Evidence ×1 Temporal Knowledge Base ×1 Ancient Chinese ×1 Document Recognition ×1 AI-generated text detection ×1 mixed-origin text ×1 content-expression routing ×1 gated composition ×1 HART benchmark ×1 Causal Recommendation ×1 Deep Twin Networks ×1 Holdback ×1 Impression Efficiency ×1 Difference-in-Differences ×1 Censored Rating ×1 BOCS ×1 Competitor-Aware ×1 Content Optimization ×1 Video Pretraining ×1 Token Dropping ×1 Block-Causal Attention ×1 SIGReg ×1 Emotion Understanding ×1 Trajectory-Aware Reliability ×1 Selective Inference ×1 algorithm design ×1 sandbox ×1 Knowledge Probing ×1 Long-Tail ×1 Divergent Knowledge ×1 Time-aware ×1 Dynamic Dataset ×1 RCCA ×1 web app generation ×1 Micro-video Recommendation ×1 Spectral Factorization ×1 Memory-constrained serving ×1 Index Construction ×1 Meta-Evaluation ×1 Human Annotation ×1 PPSR ×1 Image Search ×1 Attention Guidance ×1 Retrieval Orchestration ×1 SQL ×1 GRPO-DAPO ×1 low-cost pretraining ×1 MuonH ×1 curriculum model averaging ×1 open-recipe ×1 RTX 5090 ×1 Validation Ratchet ×1 Architecture Pattern ×1 Trust Domains ×1 Execution Audit ×1 Persona Drift ×1 weak model guidance ×1 Signals ×1 Conversion ×1 eBay ×1 Doc2Doc MT ×1 trajectory filtering ×1 segment-level loss ×1 mechanism design ×1 token-level advertising ×1 LLM auction ×1 DSIC ×1 KL-regularized welfare ×1 generation-native advertising ×1 Modality Collapse ×1 Asymmetric Fusion ×1 Pathway Isolation ×1 Visual Text Understanding ×1 root-cause attribution ×1 Topology Mask ×1 Multi-Domain Sequence ×1 Dual Query ×1 order consistency ×1 LLM scorer ×1 decision stability ×1 presentation dependence ×1 Fabricated Evidence ×1 Act-Don't-act Gate ×1 permission systems ×1 user-authored policies ×1 overreach ×1 instruction sensitivity ×1 prompt robustness ×1 layer dynamics ×1 linear probes ×1 moral foundations ×1 semantic data processing ×1 LLM cost optimization ×1 adaptive query processing ×1 Cortex AISQL ×1 Audio Question Answering ×1 Attention Routing ×1 Transduced Language Models ×1 Unbiased Estimation ×1 Beam Sampling ×1 Finite-State Transducer ×1 Probability Estimation ×1 Incremental Indexing ×1 LLM Controller ×1 Agent OS ×1 grammar-constrained decoding ×1 context-free grammar ×1 pushdown automata ×1 beam search ×1 lookahead ×1 logits ×1 Representation Bias ×1 Decoder LLM ×1 Entropy-Weighted Loss ×1 Post-Training Correction ×1 Long-form Generation ×1 Adaptive Signal Routing ×1 Cross-modal Synthesis ×1 Latent Hubs ×1 sequential sampling ×1 self-refinement ×1 Parser States ×1 PLVR ×1 Symbolic Backpropagation ×1 bias-variance ×1 Video Recommendation ×1 Text Embeddings ×1 Content Matching ×1 Interaction-Aware ×1 reasoning diversity ×1 parameter interpolation ×1 Countdown ×1 refusal ×1 linear representation ×1 unanswerable questions ×1 survival analysis ×1 repurchase prediction ×1 AFT ×1 grocery recommendation ×1 multi-horizon ×1 Symbolic Regression ×1 AutoFE ×1 Tabular Data ×1 OpenEuroLLM ×1 Modality-Interest Alignment ×1 Graph Refinement ×1 Module Routing ×1 Dataset-aware ×1 Communication-Efficient ×1 Hybrid Mamba-Transformer ×1 Layer Re-Configuration ×1 AutoResearch Agent ×1 optimizer survey ×1 matrix-aware methods ×1 schedule-free ×1 Semantic Head Specialization ×1 Agentic Artifact Creation ×1 Construction Policy ×1 Benchmarks ×1 content exploration ×1 creator economy ×1 two-sided marketplace ×1 corpus effect ×1 Mobile AI ×1 Modality Fusion ×1 Probe judge ×1 Small language models ×1 Reward model ×1 Efficient RL ×1 sampled-token ×1 vague goals ×1 critique-aware training ×1 tool-calling agent ×1 long-horizon reliability ×1 query language ×1 geo-localization ×1 recommendation bias ×1 Negotiation ×1 Deterministic Simulation ×1 Normalization ×1 Query Intent ×1 LRM ×1 Reward hacking ×1 Curriculum learning ×1 Agentic learning ×1 Agentic Video Generation ×1 Persistent Agents ×1 Perception-Centered Architecture ×1 Lifecycle Tasks ×1 Geo-localization ×1 Perception-Verification ×1 Multimodal Encoder ×1 Single-Tower ×1 Vision-Document Retrieval ×1 Conversational Grounding ×1 GitHub ×1 Reference Resolution ×1 Token-level RL ×1 Entropy-adaptive Advantages ×1 Research Planning ×1 PaperGym ×1 robust training ×1 N-gram Embedding ×1 Realistic Inputs ×1 Parameter Training ×1 Interactive Learning ×1 Grounded Theory ×1 Agent Behavior Analysis ×1 Rejection Sampling Fine-Tuning ×1 Meta-Harness ×1 knowledge gating ×1 task construction ×1 verifiable rewards ×1 rubric induction ×1 evaluation-first ×1 iterative revision ×1 scientific research ×1 criterion verification ×1 co-engagement ×1 item-to-item retrieval ×1 LLM-MLLM ×1 production ads ×1 Pointwise Reranking ×1 Score Polarization ×1 Generative Constraint ×1 Next-Basket Recommendation ×1 Feature Attribution ×1 Repurchase ×1 Frozen Index ×1 Query Encoder ×1 attention sink ×1 Product Cluster ×1 Negative Constraints ×1 Open IE ×1 fact verification ×1 span localization ×1 Remote Sensing ×1 POI ×1 Numerical Encoding ×1 Evidence-conditioned ×1 structured control ×1 protocol stability ×1 schema validation ×1 memory tokens ×1 Execution Cost ×1 Low-Rank Residual ×1 Gradient Competition ×1 FWHT ×1 Set Prediction ×1 Hungarian Matching ×1 Presence Gating ×1 ANN Query Reduction ×1 Interest Collapse ×1 State-Space Model ×1 Review Feedback ×1 Item Dynamics ×1 image captioning ×1 SimLoss ×1 fine-grained ×1 single-pass ×1 Student Simulator ×1 Sparse Data ×1 compute plateau ×1 guidance gap ×1 Cloze Test ×1 Listwise Selection ×1 teacher model ×1 Token Routing ×1 Teacher Entropy ×1 Factual Recall ×1 First Mistake Identification ×1 Governed Analytics ×1 Deterministic Policy Execution ×1 LLM Intent Parsing ×1 Tool Planning ×1 Replayability ×1 Sequential Transformer ×1 Dual-Gated Attention ×1 RoTE ×1 Shared-Prefix Serving ×1 CVR Ranking ×1 Training Data Attribution ×1 Response Rewriting ×1 SparseGPT ×1 LightGBM ×1 Event-Centric ×1 Long-Video QA ×1 Graph Machine ×1 Sparse Dynamic Routing ×1 Hungarian Algorithm ×1 Sinkhorn ×1 KGQA ×1 SPARQL ×1 Think-Act-Observe ×1 Preference Tracking ×1 Repository Mining ×1 Streaming Video Understanding ×1 Diversity-aware Retrieval ×1 Counterfactual RL ×1 Engagement Modeling ×1 training-time augmentation ×1 SASRec ×1 BERT4Rec ×1 Transformer Ranking ×1 Trajectory Reuse ×1 Semantic Routing ×1 TaaS ×1 Video Generation Evaluation ×1 Multimodal Eval ×1 Low-latency ×1 Dynamic MIPS ×1 Edge Retrieval ×1 Index-Free ×1 Partial Scan ×1 Teacher gating ×1 Keyword Matching ×1 Co-Evolving RL ×1 Declarative Attention ×1 Verbal RL ×1 Feedback Taxonomy ×1 Self-Refine ×1 Latent State Internalization ×1 LLM-Agent Collaboration ×1 Verbalization Debt ×1 Non-language Agent ×1 Chess ×1 Bayes error ×1 BER ×1 AUC ×1 soft labels ×1 isotonic regression ×1 estimator evaluation ×1 Progressive Optimization ×1 Orthogonal Purification ×1 Persistent Discovery Context ×1 Data Discovery ×1 Closed-Loop Optimization ×1 Resource Allocation ×1 answer verification ×1 multi-domain LLM ×1 Revision Propagation ×1 Conversational Artifacts ×1 trajectory balance ×1 Hybrid LLM ×1 Visual Design ×1 HTML-CSS ×1 Editable Delivery ×1 Render-and-Reflect ×1 MLLM Safety ×1 Cross-modal ×1 Representation Transfer ×1 LLM Scoring ×1 Kernel Generation ×1 TPU ×1 Compiler Feedback ×1 OOD recommendation ×1 counterfactual generation ×1 FDR control ×1 selection inference ×1 Predicted-State Matching ×1 Action Ranking ×1 SOP Engineering ×1 Multimodal Trace ×1 Professional Software ×1 conflict detection ×1 egocentric KB ×1 graph traversal ×1 Physics Understanding ×1 Omni-Camera ×1 adaptive retrieval ×1 corpus graph ×1 group-level expansion ×1 explore-then-exploit ×1 reasoning-intensive retrieval ×1 Image Forensics ×1 Multi-Tool Attribution ×1 Deepfake ×1 reflection ×1 game theory ×1 verifier grounding ×1 memory editing ×1 Spatial Perception ×1 random attention ×1 Text2Cypher ×1 test-time inference ×1 adaptive budget ×1 execution-guided refinement ×1 MLLM Embedding ×1 Reranker Distillation ×1 Rank-KL ×1 Listwise Training ×1 intrinsic dimensionality ×1 training seed ×1 model selection ×1 E5M3 block scale ×1 DocVQA ×1 Adaptive-k Retrieval ×1 Tail Optimization ×1 Generative Policies ×1 Best-of-k ×1 Advantage ×1 Conditional Computation ×1 Dual System ×1 Latent Refinement ×1 Visual Token Allocation ×1 Long-Video MLLM ×1 Frame Selection ×1 Orthogonal Matching Pursuit ×1 Average Treatment Effect ×1 Input Warping ×1 GP-UCB ×1 Kernel ×1 Pooled evaluation ×1 Retrieval model selection ×1 Incremental evaluation ×1 Cost-efficiency ×1 Cross-workspace ×1 State Coverage ×1 Multi-Teacher OPD ×1 Adversarial Examples ×1 NLIP ×1 Representational Alignment ×1 ReSO ×1 Moral Categorization ×1 Layer Dropout ×1 Stochastic Depth ×1 Self-Speculative Decoding ×1 annotation ×1 medical LLM ×1 Skill Synthesis ×1 Code Mining ×1 Entailment Distillation ×1 Answerability ×1 Abductive Recall ×1 LLM4IR ×1 pointwise reranking ×1 Outcome-Blind ×1 Video Diffusion Models ×1 Denoising Stages ×1 VBench ×1 error clustering ×1 bootstrap stability ×1 prompt bloat ×1 Banking ×1 Advantage Estimator ×1 Rule Induction ×1 Horn Clause ×1 Perplexity Filtering ×1 Sample Selection ×1 Wall-clock Efficiency ×1 shared endpoints ×1 preregistration ×1 ORPO ×1 Refusal Circuits ×1 Top-K Routing ×1 Shared Experts ×1 Oracle Risk Bounds ×1 Sparse Activation ×1 Statistical Learning Theory ×1 Hyperbolic Geometry ×1 Budgeted Verification ×1 Prompt Design ×1 Record Directive ×1 Plan Pointer ×1 Monte Carlo Rollouts ×1 Step Importance ×1 attention masking ×1 Adversarial RL ×1 simulation-based power ×1 SFT-RL-DPO ×1 Nearline-Online ×1 epistemic warrant ×1 pairwise recommendation ×1 code repair ×1 edit fidelity ×1 minimal edits ×1 Co-work ×1 AI-assisted peer review ×1 evidence-guided refinement ×1 overcritique ×1 omission ×1 selective risk control ×1 Situational Awareness ×1 Meeting Delegation ×1 Online Decision ×1 perturbation ×1 intrinsic dimension ×1 HotFlip ×1 Residual Correction ×1 Multivariate ×1 iTransformer ×1 RISE ×1 self-extrapolation ×1 Local LLM ×1 Identity Preservation ×1 Generative Image Models ×1 Subject-driven Generation ×1 Persistent Identity ×1 Temporal Self-Distillation ×1 agent construction ×1 end-to-end evaluation ×1 LLM Inference Acceleration ×1 rubric learning ×1 neural function ×1 editable state ×1 state management ×1 Context Modeling ×1 Item Posterior ×1 Blackwell ×1 LLM pre-training ×1 auxiliary views ×1 knowledge acquisition ×1 paraphrasing ×1 Financial QA ×1 Chunk Enrichment ×1 Long-document Retrieval ×1 instruction duplication ×1 procedural following ×1 trajectory state ×1 answer engineering ×1 autoregressive LM ×1 adaptive token selection ×1 pairwise relational alignment ×1 Extended Kalman Filter ×1 Model Upgrade ×1 Memory Portability ×1 Two Tower ×1 Category Adapter ×1 Co-purchase ×1 Production RecSys ×1 Agent Quorum ×1 Graph Neural Network ×1 Coherence ×1 Entity Graph ×1 multi-agent teams ×1 interchangeability ×1 coordination efficiency ×1 agent swap ×1 emergent conventions ×1 Subset Selection ×1 Soft Context Compression ×1 Assume-Guarantee ×1 GUI+CLI ×1 Environment-to-Data ×1 data policy ×1 Generative NVS ×1 Mirror Reflection ×1 Multi-view Diffusion ×1 Virtual View ×1 GCN ×1 Signal Erosion ×1 Causal Masking ×1 graph anomaly detection ×1 graph-language alignment ×1 von Mises-Fisher ×1 zero-shot transfer ×1 Matryoshka representation ×1 Graph Complexity ×1 Training-Free Optimization ×1 Inventory-Grounded ×1 AI Search ×1 Policy Guidelines ×1 Tree-based Reasoning ×1 Evidence-Intensive QA ×1 Topology-Aware Evidence Gathering ×1 Debate ×1 Clinical Diagnosis ×1 Diagnostic Safety ×1 Differentiable Measurement ×1 neurosymbolic ×1 SQL grounding ×1 AI Dependence ×1 Psychological State ×1 Classroom ×1 mHC ×1 Residual Streams ×1 LLM Architecture ×1 Model Analysis ×1 DeepSeek-V4-Flash ×1 data-synthesis ×1 execution-grounded ×1 API-agent ×1 Spatio-Temporal ×1 Neuro-Symbolic ×1 Study Platform ×1 Adaptive Edge ×1 necessity ×1 sufficiency ×1 black-box evaluation ×1 agent oversight ×1 Bitcoin ×1 Ranking Correction ×1 Convex Optimization ×1 ANN Index ×1 Cross-modal Dilution ×1 Item Embedding ×1 Trade-Up Recommendation ×1 Product Relations ×1 Retail Search ×1 Design Docs ×1 AI Coding Agents ×1 Performance Modeling ×1 Symbolic IR ×1 SymPy ×1 sampling completeness ×1 rarefaction ×1 Skill Correctness ×1 Specification Reasoning ×1 Intent Mask ×1 Sandbox Validation ×1 Skill Hierarchy ×1 Spectral Clustering ×1 Trajectory Mining ×1 Graph Contrastive ×1 Omni-modal LLM ×1 Cross-modal Conflict ×1 Perceptual vs Propositional ×1 Automated Evaluation ×1 WearableQA ×1 Health Reasoning ×1 Wearable Data ×1 Multiple Choice ×1 requirement-aware reasoning ×1 MLLM agents ×1 must-have-nice-to-have ×1 Neuro-Symbolic Reasoning ×1 human values ×1 geometry ×1 influence score ×1 Agentic Coding ×1 Execution Gate ×1 limited feedback ×1 regret minimization ×1 parallel reading ×1 Language Transfer ×1 Low-resource ×1 Seed Variance ×1 hybrid LLM ×1 Online Co-Training ×1 Context Parallel ×1 Pipeline Parallel ×1 RL Post-Training ×1 state-space model ×1 Block-Stacking ×1 3D-to-2D ×1 Bayesian compression ×1 spike-and-slab ×1 token redundancy ×1 Geometric Dynamics ×1 Soft Label ×1 Rebuttal ×1 Contrastive Facets ×1 Coverage-aware Ranking ×1 EAGER ×1 I2Q ×1 Preference-Aware Reward ×1 Spectral Compression ×1 EigenLI ×1 Omni Interaction ×1 Thinker-Talker ×1 Feedback Enrichment ×1 Graph-based Retrieval ×1 SHPO ×1 Lost-in-the-Middle ×1 TC-ECE ×1 Procedural Graph ×1 Validation Gating ×1 Megatron-LM ×1 binary hashing ×1 FAISS ×1 Pass@k ×1 Reasoning Coverage ×1 Routing Harness ×1 Scenario-Based Testing ×1 Open-Source LLM ×1 Anchor Tokens ×1 Ranking Model ×1 Task-Aware ×1 Scheming ×1 Safety Benchmark ×1 SWE-Bench Pro Verified ×1 GaugeLasso ×1 group lasso ×1 transformer compression ×1 gauge symmetry ×1 image tokenizer ×1 interactive video retrieval ×1 video browser showdown ×1 Multi-ID ×1 Passage-Level ×1 DocID Design ×1 Multi-Target Learning ×1 data mixing ×1 domain coverage ×1 controlled experiment ×1 State-Policy Distillation ×1 Annotation Expansion ×1 Functional Equivalence ×1 bias detection ×1 relative representations ×1 entity verification ×1 counterfactual stability ×1 Programmable State ×1 3D OBB ×1 Block-wise Diffusion ×1 Shapley Attribution ×1 Cooperative Game Theory ×1 tool_calling ×1 structured_output ×1 LLM_judge ×1 policy_layer ×1 Program-Solve ×1 code executor ×1 arithmetic reliability ×1 medical calculators ×1 Relevance Judgment ×1 Subcorpus Sampling ×1 Recall@1000 ×1 First-stage Retriever ×1 Differentiable Search Index ×1 NQ320K ×1 Atomic Identifiers ×1 Sketch Attention ×1 Low-Rank Caching ×1 STCA ×1 System Co-design ×1 Semantic Interface ×1 Fact-Checking ×1 LLM-as-Verifier ×1 Counterfactual Training ×1 Generator-in-the-loop ×1 Cross-encoder Reranker ×1 Next Concept Prediction ×1 Latent Space LM ×1 VideoLLM ×1 Frame Sampling ×1 Task-Agnostic Preprocessing ×1 Environment Adaptation ×1 Study Budget ×1 Frozen Solver ×1 Rare Entities ×1 On-policy correction ×1 Imitation learning ×1 IMO ×1 multi-model ensemble ×1 Bottom-Up Clustering ×1 Codebook Utilization ×1 Autoformalization ×1 VPU ×1 Z3 ×1 Fisher-Rao geometry ×1 model steering ×1 natural gradient ×1 output geometry ×1 low-bit quantization ×1 decision flip ×1 Vamana ×1 margin analysis ×1 quantization robustness ×1 TREC DL ×1 NDCG ×1 Claim Auditing ×1 Ground Truth ×1 Data Catalog ×1 Sensitivity Classification ×1 Visual-Language-Action ×1 competence gating ×1 forecast combination ×1 LLM forecasting ×1 Query-Adaptive Thresholding ×1 Hub Penalization ×1 Grounding Instruction ×1 Retrieval Recall ×1 Evaluation Methodology ×1 session co-occurrence ×1 word2vec ×1 song embeddings ×1 artist-residual ×1 Native Multimodal ×1 Encoder-free ×1 Generative Decoder ×1 synthetic rewards ×1 diagnostic reasoning ×1 digital advertising ×1 ontology ×1 temporal decay ×1 supersession ×1 confabulation ×1 Verbatim Extraction ×1 selective forgetting ×1 semantic slots ×1 rate-distortion ×1 Qualitative Coding ×1 Agent Debate ×1 Consensus ×1 Thematic Analysis ×1 LinearUCB ×1 Skill Search ×1 layerwise intervention ×1 internal knowledge retrieval ×1 causal probing ×1 request routing ×1 Decision aggregation ×1 Runtime Harness ×1 Personalized Aggregation ×1 Utility Signals ×1 Negative Self-Distillation ×1 Unlikelihood Training ×1 Label-Free Training ×1 Root-Cause Attribution ×1 Continual Search ×1 MegaRCA-Mix ×1 F1 ×1 AI benchmarks ×1 search engine ×1 benchmark database ×1 evidence tracking ×1 attention sparsification ×1 LM loss ×1 Top-K selector ×1 XAI ×1 LLM Function Calling ×1 Energy Forecasting ×1 Intent Parsing ×1 DeBERTa-v3 ×1 MC Dropout ×1 verbosity bias ×1 MultiAgent System ×1 Ethics Education ×1 Group Discussion ×1 Agentic Share-of-Search ×1 LLM-Mediated E-Commerce ×1 Generative Recommender Systems ×1 Competitive Intelligence ×1 role-play ×1 clinical training ×1 Voice Interaction ×1 MInTRL ×1 off-policy intervention ×1 advantage regression ×1 training schedule ×1 target leakage ×1 glass-box models ×1 fairness audit ×1 Retrieval Sufficiency ×1 Distribution Shape ×1 Vision Document Retrieval ×1 Hierarchical Organization ×1 NVMe ×1 open-source LLMs ×1 artificial drive ×1 persistent state ×1 self-adaptation ×1 control ×1 SpeechLLM ×1 streaming reasoning ×1 Attention Mining ×1 news framing ×1 fact preservation ×1 framing inversion ×1 controlled test ×1 Routing Stability ×1 Content Risk Control ×1 Continued Pretraining ×1 Policy Internalization ×1 Backdoor Detection ×1 Model Safety ×1 Discrete Autoregressive ×1 Progressive Pretraining ×1 music generation ×1 audio tokenizer ×1 context segmentation ×1 similarity-contractive window ×1 Time-Critical QA ×1 Overlapping-Evolving Documents ×1 FKL ×1 RKL ×1 per-token gating ×1 distance generalization ×1 synthetic tasks ×1 Cascaded Recommender Systems ×1 Cross-Stage Optimization ×1 Multi-Task Fusion ×1 Attribute Bias Regularization ×1 structured documents ×1 experience edges ×1 adaptive escalation ×1 multi-round retrieval ×1 Counteraction ×1 LM-head geometry ×1 Generalized Agent Iteration ×1 Generalized Policy Iteration ×1 Formal Framework ×1 Edge-Cloud Agent ×1 Visual-Need Router ×1 Inference-time Compute ×1 Kashin decomposition ×1 DCT ×1 RL for LLMs ×1 dynamic sampling ×1 curriculum ×1 Matthew Effect ×1 MDP mid-training ×1 attention quantization ×1 TabPFN ×1 Curriculum RL ×1 Exploration Bottleneck ×1 Tool-use ×1 Closed-Loop Benchmark Synthesis ×1 Artifact Dependency Graph ×1 Verification & Repair ×1 Skill Orchestration ×1 Cross-Modal ×1 Emergence ×1 concept representation ×1 affine subspace ×1 multinomial concepts ×1 Low-Rank Adaptation ×1 Behavior Quotient ×1 Decision Distribution ×1 drift-constrained optimization ×1 layer-selective tuning ×1 QA-only fine-tuning ×1 multilingual translation ×1 NVIDIA Hopper ×1 GMMA ×1 decode ×1 benchmark leakage ×1 truthfulness ×1 surface features ×1 dataset cleaning ×1 LLM eval ×1 Audit-Prune ×1 speech interaction ×1 asynchronous delegation ×1 realtime agent ×1 Prefetch Scheduling ×1 Hierarchical Solver ×1 Large-scale Optimization ×1 KKT ×1 Edge Localization ×1 Real-time Speech ×1 Think-While-Speaking ×1 Full-Duplex ×1 specialist distillation ×1 reasoning trajectories ×1 rankECE ×1 probabilistic forecast ×1 procedural reasoning ×1 subquadratic attention ×1 heterogeneous disaggregation ×1 Capability Composition ×1 Tilted-Target DOPD ×1 Checkpoint Offload ×1 Optimizer Offload ×1 Symbolic Policy Learning ×1 Self-directed Inquiry ×1 Intermediate Verification ×1 Spectral Reparameterization ×1 Long Context Extension ×1 Learning Rate Scaling ×1 Verifiable Experience Pipeline ×1 Prompt Scaffolding ×1 Exploration Potential ×1 Training Data Refinement ×1 exploration policy ×1 replay simulator ×1 off-policy evaluation ×1 Discovery Foundation Models ×1 Open-Ended Discovery ×1 Research State ×1 Verification Gating ×1 Discovery Skill Evolution ×1 Embodied Navigation ×1 Spatiotemporal Graph ×1 Tool Interface ×1 Guard Model ×1 Execution Grounded ×1 Register Tokens ×1 Bounded-State Reasoning ×1 Chunked SFT ×1 theory of mind ×1 privileged distillation ×1 human-aware LLM ×1 mental state ×1 Harness Self-Improvement ×1 Modular Evolution ×1 Contrastive Trajectory ×1 Benchmark-Disjoint ×1 Mixture-of-Memory ×1 Memory Embeddings ×1 Streaming Multimodal ×1 Evidence Verification ×1 Audio-Visual Reasoning ×1 Answer Gating ×1 low-bit attention ×1 kernel optimization ×1 Orthrus ×1 Losslessness ×1 Weight Editing ×1 Refusal Feature Ablation ×1 Post-Hoc Defense ×1 Policy Mirror Descent ×1 Critic-free ×1 Citation Validation ×1 Citation Laundering ×1 Counterfactual Defense ×1 Linear Projection ×1 Product of Experts ×1 Adversarial Stress Testing ×1 Long-Horizon Evaluation ×1 Semantic Layer ×1 top-k decoding ×1 Drug Repurposing ×1 Population-based Search ×1 Relevance Estimation ×1 Text-as-Policy ×1 Trading ×1 Generative Pre-training ×1 Sparse-Dense Transfer ×1 Retrieval Abstention ×1 Score Distribution ×1 Magnitude Mirage ×1 BRP@k ×1 MRR@k ×1 stochastic retrieval ×1 Social Reasoning ×1 User-Mediated ×1 Trial vs Reorder ×1 Skill Repair ×1 Retest-Conditioned Update ×1 Obfuscated text ×1 Web agent ×1 Risk detection ×1 Evidence-grounded ×1 Factual Decoding ×1 Internal Attribution ×1 Interactive Workspace ×1 Evolution Conflict ×1 KV Memory ×1 Headline Generation ×1 Explore-Exploit ×1 Theorem Proving ×1 Tree Aggregation ×1 TCS-Bench ×1 midtraining ×1 selective generalization ×1 neologism ×1 experience bank ×1 shared-memory ×1 Git ×1 autonomous-research ×1 diversity-aware-selection ×1 BI ×1 Agentic Tools ×1 Swing ×1 i2i retrieval ×1 approximate algorithms ×1 top-K retrieval ×1 graph algorithms ×1 Model Serving ×1 Feature Store ×1 Autotuning ×1 Graph World Model ×1 Diversity Capping ×1 Retrieval-Stage Diversification ×1 Online A-B Testing ×1 Parameter Tuning Sequence ×1 E-commerce Recommendation ×1 Comparison Oracles ×1 Likelihood Displacement ×1 Context Compiler ×1 Constrained Controller ×1 World Reasoning ×1 Routing Prediction ×1 Memory Wall ×1 teacher bias ×1 reward feedback ×1 Panoptic Grounded Captioning ×1 Mask Proposal Selection ×1 Dense Captioning ×1 Pixel-level Grounding ×1 reference-guided ×1 SWE ×1 interactive web apps ×1 replay verification ×1 Region-Level Retrieval ×1 Large Multimodal Models ×1 Region-Aware Encoder ×1 Prompt Routing ×1 Weak Risk ×1 Graph Token ×1 Code Repair ×1 Predictive Coding ×1 GPU inference ×1 batch scheduling ×1 Trajectory Probing ×1 Answer Quality Prediction ×1 harness design ×1 action space ×1 ablation study ×1 incremental memory ×1 NPC ×1 recurrent-attention ×1 Ensemble Pruning ×1 Mahalanobis Distance ×1 EOS mismatch ×1 semantic EOS ×1 special tokens ×1 Tool Call ×1 Progress Reporting ×1 Deductive Reasoning ×1 Possibility Matrix ×1 Game Environment ×1 Normative QA ×1 Scope Resolution ×1 Production Evaluation ×1 task vectors ×1 ethical alignment ×1 task arithmetic ×1 Llama ×1 Recursive Depth ×1 Model Growth ×1 Boundary Operator ×1 Expert Pruning ×1 Second-Order ×1 Adaptive Teacher Retirement ×1 Teacher Construction ×1 AI audit ×1 ConsumerQ ×1 Language Agent ×1 Dual-Process ×1 Memory Module ×1 ScienceWorld ×1 Action Fusion ×1 privileged-information ×1 cross-mode-transfer ×1 spatial intelligence ×1 MLLM benchmark ×1 active perception ×1 metric control ×1 Degradation Synthesis ×1 Diagnostic Clustering ×1 Hybrid Reasoning ×1 Bayesian Update ×1 Weight Generation ×1 Region-level RL ×1 Fine-grained perception ×1 Visual token compression ×1 Sparse encoding ×1 Proposal network ×1 tool routing ×1 multilingual-e5 ×1 critic learning ×1 value flattening ×1 sparse supervision ×1 Self-Evolving Index ×1 Index Optimization ×1 On-Demand Attention ×1 local attention ×1 AI Agent Security ×1 Payment Authorization ×1 Open Agent Passport ×1 Observation Supervision ×1 record-replay ×1 regression-testing ×1 CI-CD ×1 RL environments ×1 source code ×1 MiMo ×1 logit lens ×1 causal attribution ×1 Document VQA ×1 Answer Grounding ×1 AR-to-DLM ×1 drift correction ×1 Evidence Refinement ×1 harm laundering ×1 representational harm ×1 toxicity metrics ×1 GPT lineage ×1 Relational Attention ×1 BabyLM ×1 Dual Attention Transformer ×1 NextLat ×1 Copy Detection ×1 expert composition ×1 block-level routing ×1 hypernetwork ×1 serving efficiency ×1 AI evaluation ×1 claim safety ×1 closed-loop ×1 parallelism ×1 uniform diffusion ×1 Gaussian diffusion ×1 critical windows ×1 Overclaiming ×1 Feedback Attribution ×1 Trajectory Normalization ×1 Obstacle Avoidance ×1 UI Agent ×1 Digital Environments ×1 audio-visual dialogue ×1 multi-agent data synthesis ×1 Qwen3-Omni ×1 on-policy fine-tuning ×1 privileged teacher ×1 policy distillation ×1 end-to-end driving ×1 3DGS simulation ×1 Troubleshooting Agent ×1 Timeline ×1 Case Retrieval ×1 Hybrid Agents ×1 Execution-Grounded Reward ×1 representational adequacy ×1 agents ×1 Semantic Parsing ×1 Schema Alignment ×1 Social Media Metaphor ×1 Graph-structured Skill ×1 inference-engine ×1 LLM-security ×1 sandbox-escape ×1 agentic-harness ×1 Subword Tokenization ×1 Lossless Decomposition ×1 Orthographic Variation ×1 Vocabulary Efficiency ×1 GPT-2 ×1 disaggregated evaluation ×1 prediction-powered inference ×1 small area estimation ×1 Bayesian smoothing ×1 cross-validation ×1 Teacher-Student Discrepancy ×1 Abstract Token Curriculum ×1 continuous thoughts ×1 alignment midtraining ×1 instruction fidelity ×1 conversational agent ×1 randomized trial ×1 Product Retrieval ×1 Temporal Localization ×1 Livestream E-commerce ×1 Multi-perspective Reasoning ×1 Text Ranking ×1 multi-view memory ×1 jailbreak ×1 defense combination ×1 Membership Inference ×1 Pretraining Data Detection ×1 Entropy Correction ×1 Free Energy ×1 latent chain-of-thought ×1 Continual Adaptation ×1 Behavior Diagnosis ×1 confidence scoring ×1 Best-of-n ×1 Inference-time Alignment ×1 Speculative Inference ×1 Exponential Noise ×1 Robotic Control ×1 Discrete Actions ×1 AdaGrad ×1 Matrix Optimization ×1 Online Mirror Descent ×1 Adaptive Methods ×1 Deep Learning Training ×1 Memory Decision ×1 Zero-parameter ×1 Speculative Sampling ×1 Poisson Process ×1 Drafter Invariance ×1 Multi-Draft ×1 Certification ×1 Exact Binomial ×1 Deployment ×1 L0-regularization ×1 LLM acceleration ×1 sparse MoE ×1 Text-and-Tabular ×1 Automated Feature Discovery ×1 full-duplex speech ×1 speech-to-speech ×1 streaming architecture ×1 educational LLM ×1 capability-oriented ×1 gating ×1 noise filtering ×1 Scalar Drift ×1 dynamic tree ×1 stochastic sampling ×1 lossless acceleration ×1 Training-free Proxy ×1 Random Matrix Theory ×1 LLM Pruning ×1 feature superposition ×1 navigation ×1 Clinical Trial ×1 Regulatory Science ×1 Retrieval-aware RL ×1 Interactive Retrieval ×1 internal probing ×1 sandbagging ×1 unlearning verification ×1 concealed information test ×1 Teamwork Strategy ×1 Demonstrability ×1 Reward-guided Decoding ×1 Category-Aware RL ×1 Iterative Expert Training ×1 Policy Integration ×1 Software Engineering Agents ×1 Gradient Estimation ×1 Sparse Distillation ×1 Agent Continual Learning ×1 Multi-stage Post-training ×1 LoRA Experts ×1 SDFT ×1 LLM Survey ×1 Cognitive Agent ×1 Personalized Companion ×1 Longitudinal Support ×1 Tool-calling agents ×1 Synthetic data generation ×1 Edge cases ×1 Finetuning ×1 Harness optimization ×1 Policy compliance ×1 harness distillation ×1 agent-as-harness ×1 trajectory learning ×1 System-One-Two ×1 Adaptive Stopping ×1 token-level correction ×1 data annotation ×1 Omni-modal Embedding ×1 Focal Loss ×1 Low-rank Decomposition ×1 SVG Generation ×1 Vision-Language Judge ×1 VABench ×1 PDF Normalization ×1 Markdown Conversion ×1 Enterprise Knowledge ×1 3D-aware memory ×1 camera control ×1 long-horizon consistency ×1 Latent Decoding ×1 Event-Structured ×1 AI Research Agent ×1 Self-Modification ×1 collusion ×1 Kimi Delta Attention ×1 Critical-State RL ×1 nested sampling ×1 Self-Evolving Policy ×1 Task Completion ×1 Cost-Latency ×1 Exposure Accounting ×1 Copy Ceiling ×1 economic decisions ×1 Matrix Factorization ×1 Cross-Layer Redundancy ×1 Structured Sparsity ×1 cascade ×1 evaluation cost ×1 reward models ×1 ICL ×1 retrieval vs generation ×1 learned memory ×1 lightweight routing head ×1 Computer Use Agents ×1 Subgoal ×1 text-only ×1 API ×1 Generative Agents ×1 Human-AI Co-evolution ×1 Longitudinal Simulation ×1 Counterfactual Intervention ×1 Weight Quantization ×1 Low-bit ×1 GEMV ×1 Codebook Training ×1 tree-structured speculation ×1 DeepSeek-V4 ×1 compressed attention ×1 small-world network ×1 network centrality ×1 risk ideation ×1 participatory AI ×1 multi-agent brainstorming ×1 taste ×1 decision fork ×1 Continual Document Authoring ×1 ChangeLog ×1 Claim Extraction ×1 AI Tutoring ×1 Learning Gains ×1 StudentBench ×1 tool-call ×1 serving stack ×1 Ollama ×1 concept generation ×1 search policy ×1 Annotation Budget ×1 Sequential Decision Making ×1 Simulation Benchmark ×1 Runtime Policies ×1 Action Denial ×1 temperature sampling ×1 trajectory memory ×1 x0-prediction ×1 Read-time Curation ×1 Task-Adaptive ×1 AR verification ×1 Low-bit Quantization ×1 Quantization-Aware Distillation ×1 Exposure Bias ×1 Mixed-Granularity ×1 Potential-Based Reward Shaping ×1 reasoning extraction ×1 Failure-Guided Learning ×1 Context Efficiency ×1 Selective Control ×1 Proximity Trap ×1 LYRA ×1 Multi-party dialogue ×1 Dual-track memory ×1 Speaker attribution ×1 greedy decoding ×1 FP16 ×1 LM head ×1 Dual-Mode CoT ×1 Agent Tool Calling ×1 Experiment Design ×1 Gaussian Process ×1 Code Equivalence ×1 Numerical Inference ×1 model pruning ×1 label-free ×1 Theorem Discovery ×1 Proof Difficulty ×1 Interestingness Metric ×1 Self-Expanding Library ×1 Formal Math ×1 Object Permanence ×1 Elo Evaluation ×1 TAMP ×1 LLM planning ×1 trajectory fine-tuning ×1 action prediction ×1 search control ×1 distributional serving ×1 short-video recommendation ×1 duration bias ×1 event-time distribution ×1 spin-glass ×1 optimization landscape ×1 log-depth ×1 recurrent ×1 language-modeling ×1 balanced-tree ×1 memory efficiency ×1 Entropy Calibration ×1 Deep Search Agent ×1 Role Decoupling ×1 Long-Horizon RL ×1 Summarization ×1 Planner-Synthesizer ×1 Neuron Analysis ×1 Paraphrase ×1 Bi-encoder ×1 Memory Attention ×1 Token Memory ×1 Value Projection ×1 CPU Offloading ×1 Quantile Balancing ×1 Router Gradient ×1 Causal Shortcuts ×1 Masking Strategy ×1 One Transformer ×1 Cascade Unification ×1 Sparse MoE ×1 Sequence-Native Training ×1 LLM-Evaluation ×1 Turkish-NLP ×1 Document-QA ×1 Real-time Control ×1 Human-AI Teaming ×1 Model-Harness Co-evolution ×1 CARE ×1 Difficulty Filtering ×1 autonomous ML research ×1 embedding learning ×1 Action Rehearsal ×1 representation invariance ×1 permutation SNR ×1 order sensitivity ×1 Dual-Hypergraph ×1 Knowledge Aggregation ×1 Temporal Aggregation ×1 consumer behavior ×1 bounded rationality ×1 pricing cues ×1 tool calls ×1 information search ×1 Temporal IR ×1 Temporal Embeddings ×1 Gaussian moment matching ×1 tool-calling agents ×1 enterprise data ×1 tensor decomposition ×1 Tucker ×1 low-rank approximation ×1 linear superposition ×1 guided decoding ×1 prompt enhancement ×1 semantic consistency ×1 Online Linear Programming ×1 Stochastic Gradient Descent ×1 Resource Pricing ×1 Logarithmic Regret ×1 First-order Method ×1 reverse auction ×1 cost estimation ×1 quality threshold ×1 marketplace ×1 Spatial Transformer ×1 Decentralized Control ×1 Reasoning Space ×1 Quantization Price ×1 Configuration Selection ×1 EWC ×1 SCFF ×1 feature folding ×1 memory-efficient inference ×1 training-free inference ×1 RLCD ×1 Alignment failures ×1 Zero-shot detection ×1 Calibrated decisions ×1 Action Judge ×1 State Revision ×1 EditAct ×1 CCE ×1 synthetic personas ×1 LoRA-SFT ×1 over-doom bias ×1 memory maintenance ×1 budgeted routing ×1 provenance preserving ×1 multi-fidelity bandits ×1 Alignment Illusion ×1 Representational Similarity ×1 Principal Angle Gap ×1 Multimodal Interpretability ×1 Answer Revision ×1 Decision Policy ×1 Query Classification ×1 AgentX-Model ×1 Experiment Replay ×1 Strategy-aware ×1 Multi-stage ×1 Attention Regularization ×1 Probability Mass Concentration ×1 action credit ×1 policy drift ×1 credit transport ×1 decision gate ×1 Structured Action Space ×1 Social Context ×1 Conspiracy Detection ×1 Illocutionary Force ×1 Social Media ×1 GRAFT ×1 step-level advantage ×1 trajectory graph ×1 Bellman iteration ×1 Regulatory Compliance ×1 EFSM ×1 Workflow Compilation ×1 Trace Alignment ×1 SkillOpt ×1 slate recommendation ×1 noisy scores ×1 Block Sparse ×1 Log-Linear Complexity ×1 Top-K Selection ×1 Verification Gates ×1 PyTorch Inductor ×1 Cross-Country ×1 Code-Mixing ×1 Semantic Codebook ×1 behavioral assays ×1 agent instrumentation ×1 model drift ×1 Turing Machine ×1 rank stability ×1 prompt structure ×1 trace integrity ×1 CVAE ×1 Semantic Fusion ×1 KL Annealing ×1 Failure-Driven Refinement ×1 Pareto Top-N ×1 Low-Rank Compression ×1 Many-shot ICL ×1 policy composition ×1 reward interpolation ×1 evasion ×1 Agentic Programming ×1 Trajectory Optimization ×1 Demonstration ×1 Topological Guidance ×1 Hyperbolic Embedding ×1 Algebraic Sparsification ×1 Numerical Embeddings ×1 Contextual Response ×1 Low-Rank Dynamics ×1 Feature Representation ×1 Search Evaluation ×1 Listwise Modeling ×1 LLM Assessor ×1 Reasoning-Intensive Retrieval ×1 Entropy Regularization ×1 Cross-Entropy ×1 Verifiable Tasks ×1 Search-Aware RL ×1 GR-DPO ×1 Production Search ×1 Cross-Task Generalization ×1 Covering Numbers ×1 Softmax Attention ×1 Oracle Routing ×1 NV-Embed ×1 problem discovery ×1 synthetic generation ×1 SAT solver ×1 intuition ×1 Riemannian Flow Matching ×1 Late Interaction DiT ×1 Tool-boundary Mocking ×1 Fact Checking ×1 audio language models ×1 phonological features ×1 cross-modal representation ×1 multilingual speech ×1 Game Arena ×1 Strategic Reasoning ×1 pre-logit steering ×1 test-time alignment ×1 Contextual Robustness ×1 Adversarial Context ×1 Action-oriented ×1 Temporal Applicability ×1 Tensor Parallelism ×1 Collective Coverage ×1 state folding ×1 Indexer ×1 Top-K ×1 Lossless Pruning ×1 Multi-agent scaling ×1 LLM ensemble ×1 Disjunctive tasks ×1 Compensatory tasks ×1 Plurality voting ×1 Fermi estimation ×1 MLLM serving ×1 micro-batching ×1 goodput ×1 Multi-agent debate ×1 Provenance fidelity ×1 NLI auditing ×1 Consensus synthesis ×1 Long-video Understanding ×1 Subspace Projection ×1 Head Selection ×1 Learnable Decay ×1 OPTQ ×1 stochastic OPTQ ×1 Structured Extraction ×1 Choice Model ×1 Preference Recovery ×1 Price Sensitivity ×1 Booking ×1 Rotation ×1 Grouped Asymmetric INT4 ×1 Null-Space Alignment ×1 Householder ×1 Memory efficiency ×1 Graph Sparsification ×1 Support Graph Theory ×1 Dilation-Congestion ×1 Template Generation ×1 Planner-Evaluator ×1 Visual Generation ×1 Feedback Loop ×1 confidence training ×1 inference cost ×1 X-Tree ×1 experience tokenization ×1 Evidence Summarization ×1 Read-only Expansion ×1 State Prediction Error ×1 Value Guidance ×1 Soft Tokens ×1 LOHA ×1 ACD ×1 Context Projection ×1 Stopping Rules ×1 Capped Failures ×1 Resource Aggregation ×1 self-training ×1 strategic diversity ×1 Need Graph ×1 Decision Alignment ×1 context learning ×1 Declarative Specification ×1 Data Video ×1 Data Storytelling ×1 Skill2Env ×1 Agent Post-training ×1 Difficulty Patterns ×1 Iterative Task Hardening ×1 Deployment Traces ×1 Personal GUI Agents ×1 Experience Activation ×1 Multimodal Embeddings ×1 Evidence Generation ×1 Contrastive Retrieval ×1 Boundary Readout ×1 quantized softmax ×1 straight-through estimator ×1 calibration gradients ×1 multi-agent consultation ×1 Bayesian memory ×1 reliability estimation ×1 worker routing ×1 Domain Normalization ×1 Assignment Repair ×1 Candidate Completion ×1 Catalog Update ×1 Gated Attention ×1 Fractional Dynamics ×1 Long Sequence Modeling ×1 Power-law Memory ×1 Selective State Space ×1 auditing ×1 statistical licensing ×1 Selection vs Extraction ×1 Typed Decision Model ×1 Memory Verification ×1 Distributed Evidence ×1 ResearchSpec ×1 Long-form Report ×1 off-policy distillation ×1 strong-to-weak distillation ×1 Organizational Agent ×1 Multi-User LLM ×1 Constraint-Aware Execution ×1 retrieve-rerank-generate ×1 intent detection ×1 multi-turn modeling ×1 Document Reranking ×1 Cross-Modal Distillation ×1 Long-Document ×1 Off-Policy Learning ×1 Importance Weight ×1 activation manifold ×1 Alpha-Stabler ×1 small reasoning models ×1 knowledge bottleneck ×1 selective querying ×1 cost-aware RL ×1 Adaptive computation ×1 AIME ×1 Long-Horizon Reflection ×1 Preference Drift ×1 Prompt Sensitivity ×1 Identity Cue ×1 weight orthogonalisation ×1 Pearson correlation ×1 advantage normalization ×1 Simplex ×1 Information Collapse ×1 Non-Factorized ×1 Critic Stability ×1 Variance Weighting ×1 Overlong Filtering ×1 collective learning ×1 World Knowledge ×1 Goal Diversity ×1 Contrastive Ranking ×1 architecture search ×1 LLM-driven evolution ×1 skill library ×1 GUI-CLI Orchestration ×1 Action Space ×1 Runtime Context ×1 Memory-Gym ×1 System 1-2 ×1 RCP-nDCG ×1 Item Response Theory ×1 Reranker Evaluation ×1 representation residual ×1 RL teacher ×1 hidden-state distillation ×1 Attribute Value Extraction ×1 Pseudo-Value Correction ×1 Self-Consistency Refinement ×1 Soft Memory ×1 Examination-Relevance Decomposition ×1 Search Optimization ×1 Confidence Gating ×1 Workflow Evaluation ×1 Attention Sharing ×1 Routing Metrics ×1 Maximal coupling ×1 Teacher-guided rollout ×1 Speculative decoding ×1 Tensor Product ×1 Recall ×1 Endogenous Misalignment ×1 CoT Monitoring ×1 telescopic language models ×1 elastic inference ×1 stochastic prefix supervision ×1 nested capacity ×1 Language Drift ×1 Single-Step ×1 Delta-Rule ×1 Serving Memory ×1 Pipeline Search ×1 LLM-guided AutoML ×1 Privileged Self-Distillation ×1 Hidden State Alignment ×1 Hindsight Skill ×1 Outcome-only RL ×1 Process supervision ×1 Generative verifier ×1 Chain-of-thought ×1 Cross-Harness Adaptation ×1 Graph-based ANNS ×1 LLM-Guided Pruning ×1 harness learning ×1 code revision ×1 context following ×1 long-chain ×1 Executable Harness ×1 Cost-Aware Evolution ×1 Active Investigation ×1 Tool-Use Post-Training ×1 Execution-Driven Evolution ×1 RL Credit Assignment ×1 WUSH ×1 Data-Adaptive Transform ×1 KV cache transfer ×1 heterogeneous models ×1 prefill-free ×1 AVLLM ×1 source grounding ×1 OPSFT ×1 parameter update direction ×1 Pseudo-query ×1 Recall Optimization ×1 Advisor Model ×1 Data Intelligence ×1 Data-centric AI ×1 Semifactual Interventions ×1 Harness Construction ×1 JVP ×1 Continuous Embedding ×1 Generative Model ×1 Demonstration Pack ×1 Rubric-RL ×1 ProRubric ×1 Protocol-Level Rubrics ×1 Neighborhood OPSD ×1 Expert Pool ×1 GraphForge ×1 Working Agents ×1 Multi-Head Prediction ×1 Progressive Deployment ×1 Serving Optimization ×1 Jev-style Models ×1 Compositional Action Spaces ×1 Parallel Action Spawning ×1 Branch Selection ×1 PromptReps ×1 decision-oriented ×1 recurrent state ×1 INT8 ×1 outlier compensation ×1 Legal LLM ×1 Plan Execution ×1 Automatic Prompt Optimization ×1 Generative Enrichment ×1 Multi-Image ×1 Pivotal Mistake ×1 Recovery Distillation ×1 function-aligned quantization ×1 Reward Aggregation ×1 Advantage Energy ×1 Density-Aware ×1 Recommendation Agent ×1 OCR Perception ×1 Dynamic Memory ×1 Iterative Self-Distillation ×1 Collapse Mitigation ×1 Self-improving agents ×1 Reward-free search ×1 Agent search ×1 Agentic Meta-Reasoning ×1 Artifact Graph ×1 Agent Workloads ×1 procedure adherence ×1 banking ×1 action-oriented ×1 Selective Reliance ×1 Counterfactual SFT ×1 Reject Sampling ×1 co-cheating ×1 cross-fitting ×1 Counterfactual Credit Assignment ×1 Document-level IE ×1 Potential-based Reward Shaping ×1 power sharpening ×1 parallel tempering ×1 inference-time sampling ×1 Pairwise Verification ×1 Automated Curriculum Learning ×1 Non-stationary Bandit ×1 Failure-Pattern Arms ×1 Adaptive Exploration ×1 retrieval-based decoding ×1 Visual World Model ×1 continuous diffusion ×1 structured reasoning ×1 Egocentric Video QA ×1 Belief State ×1 Recovery ×1 Batch Size Schedule ×1 Random Features ×1 Intrinsic Time ×1 Task Taxonomy ×1 Interaction Analysis ×1 Agent Harness Evolution ×1 Topic Modeling ×1 Literature-Guided Search ×1 Module Crossover ×1 conversational recommendation ×1 synthetic benchmark ×1 label validity ×1 exact-item retrieval ×1 training augmentation ×1 nDCG@20 ×1 Item Collision ×1 Industrial Ads ×1 Prompt Variations ×1 Selective Observation ×1 Evidence-Driven ×1 WindowsAgentArena ×1 Sharpening Tax ×1 pass@K ×1 PTGS ×1 Multi-hop Search ×1 Multi-Agent Evaluation ×1 Disagreement Resolution ×1 Rubric Learning ×1 Cybersecurity ×1 Multi-task SFT ×1 Competition ×1 latent diffusion ×1 generative perplexity ×1 multi-turn evaluation ×1 conversational capture ×1 Pólya urn ×1 SigLIP ×1 proactive generation ×1 stateful inference ×1 memory layer ×1 exact match ×1 Context Curation ×1 Catalog Search ×1 SMB ×1 LLM user profiling ×1 context-aware recommendation ×1 streaming recommendation ×1 consumption regime ×1 generative AI search ×1 AI overview ×1 news consumption ×1 WER ×1 Spatial guidance ×1 Synthetic data ×1 Visual perception ×1 Self-training ×1 Label-Free ×1 In-Flow ×1 Agent Creation ×1 Fault Attribution ×1 Behavioral Cues ×1 Hierarchical Supervision ×1 Multi-Resolution Relevance ×1 low-rank compression ×1 orthogonal projection ×1 Data Hardening ×1 Serving-time Gate ×1 VideoLLMs ×1 Activation Injection ×1 Agent Web ×1 Evidence Fusion ×1 Proactive Compaction ×1 Personalization Encoder ×1 State Repair ×1 Frozen Host ×1 Temporal Correction ×1 Preference State ×1 sequence masking ×1 response filtering ×1 Patent Search ×1 Biaffine Attention ×1 Invention Graph ×1 Long Document Retrieval ×1 Graph Transformer ×1 optimizer analysis ×1 BF16 precision ×1 top-k KL ×1 Schema Induction ×1 User-Generated Content ×1 Source Learning ×1 Source Model ×1 Self-Directed Learning ×1 Task-Guided Learning ×1 Terminal Harness ×1 GUI替代 ×1 Reasoning Language Alignment ×1 German QA ×1 LLM Generative Recommendation ×1 Instruction-Guided ×1 Counterfactual Margin ×1 History Bias ×1 Asymmetric Loss ×1 Trajectory Rewriting ×1 Multi-Harness ×1 low-precision inference ×1 mixed-precision ×1 Adversarial Distillation ×1 SDXL ×1 Hierarchical Scoring ×1 Semantic Communication ×1 Multi-Robot Coordination ×1 Distributed Systems ×1 Latent Structuring ×1 Routed Optimization ×1 Geometry Alignment ×1 table retrieval ×1 join graph ×1 Local Support Learning ×1 Counterfactual Auditing ×1 Clinical Triage ×1 Pointwise Value Matching ×1 Sparsification ×1 Soft Top-K ×1 Per-Term Thresholding ×1 FLOPs Regularization ×1 Latent JEPA ×1 Chemical Reasoning ×1 Joint Embedding Prediction ×1 Molecular Optimization ×1 Source Preference ×1 End-to-End Search ×1 KL anchor ×1 Associative Retrieval ×1 Query Encoder LoRA ×1 LLM Life Simulator ×1 Entity-aware Retrieval ×1 Non-Destructive ×1 Retrieval Fusion ×1 Organizational Agents ×1 Vacuous Credit ×1 Evidence-aware Reward ×1 PubMedQA ×1 Router Alignment ×1 Continual Pretraining ×1 Small Open Models ×1 Loop Detection ×1 Prefill Budget ×1 Local-First ×1 Complementary Visual Reasoning ×1 expert encoder ×1 iterative distillation ×1 Omni-LLM ×1 Multi-turn Tool Use ×1 Structural Understanding ×1 Causal Knowledge ×1 Quasi-Newton ×1 Kronecker-factored ×1 Recall Certification ×1 optimizer state compression ×1 TACO ×1 Graph Partition ×1 DMoN ×1 Multimodal Model ×1 Hidden State Probing ×1 Cross-Model ×1 Span-Level ×1 Internal Representations ×1 Model Collapse ×1 Entropy Rate ×1 Data Filtering ×1 Iterative Fine-Tuning ×1 self-repair ×1 ablation ×1 affine law ×1 Zero-Order Optimization ×1 Step-size Selection ×1 Adaptive Optimization ×1 forgetting ×1 style drift ×1 theory ×1 posttraining ×1 LLM-as-verifier ×1 Model specs ×1 Inconsistency detection ×1 Effective Training Time ×1 PyTorch 2 ×1 Checkpointing ×1 Parameter-Efficient Fine-tuning ×1 Selective Fine-tuning ×1 Prompt Revision ×1 Reward Composition ×1 Preference Reward ×1 Step-level credit assignment ×1 Advantage weighting ×1 Variance reduction ×1 Confidence Control ×1 Evidence Filtering ×1 error repair ×1 reasoning mismatch ×1 CoT faithfulness ×1 monitorability ×1 efficient reasoning ×1 length pressure ×1 horizon loss ×1 learning budget ×1 cross-lingual retrieval ×1 multilingual encoders ×1 selective reconstruction ×1 Gradient-based ×1 Output-layer Gradients ×1 MRVQ ×1 Residual Vector Quantization ×1 Vector Compression ×1 SFT Initialization ×1 Response Masking ×1 multi-LLM collaboration ×1 committee ×1 vector retrieval ×1 query encoder ×1 statistical query ×1 learnability ×1 evidence ×1 investigation ×1 Skill Premium ×1 Latent Recurrence ×1 Evaluation budget ×1 Generalizability theory ×1 Repeated sampling ×1 Retrieval-feedback comparison ×1 Document Expansion ×1 Relevance Filter ×1 Sequential Structure ×1 Behavioral Simulation ×1 Markov ×1 Post-retrieval Adaptation ×1 Recursive Agents ×1 Self-Evaluation ×1 Agent Oversight ×1 Behavior Graph ×1 Trace2Env ×1 Agentic World Model ×1 Environment Simulation ×1 Stateful Simulation ×1 Learning-Free ×1 token cues ×1 base model ×1 CutBCE ×1 BCE loss ×1 TPU kernel ×1 Pallas ×1 Large Vocabulary ×1 massive activations ×1 attention sinks ×1 activation magnitude ×1 Low-precision ×1 Learner-Sampler Mismatch ×1 TRIAGE ×1 DRO ×1 Query Suggestion ×1 Supply-Aware ×1 Harness Co-Evolution ×1 Task Generation ×1 Long-Context Modeling ×1 Query-Dependent Routing ×1 Chunk-Wise Attention ×1 RL-only ×1 Teacher Retirement ×1 Counterfactual Data ×1 memory self-evolution ×1 capability-driven ×1 executable memory programs ×1 intent tracking ×1 System-1 ×1 Dual-System ×1 Tool Refusal ×1 power distribution ×1 SMC ×1 looped LLM ×1 fixed point ×1 depth prior ×1 orthogonal injection ×1 Open-ended Scientific Discovery ×1 Supervisor ×1 Meta Reflection ×1 Tree Routing ×1 Sparse Models ×1 agentic retrieval ×1 cost ×1 evidence-to-action ×1 Model-Agnostic ×1 SWE Agent ×1 Failure-Conditioned ×1 Negative Policy ×1 Rollout Policy ×1 Programmatic Search ×1 Evidence Presentation ×1 Candidate Workspace ×1 Search Interface ×1 critic-free ×1 Shared Vocabulary ×1 Gradient Diagnostics ×1 PAMO ×1 MediateRec ×1 Memory-efficient ×1 Low-rank gradient ×1 KL control ×1 LLM-enhanced recommendation ×1 long-short-term memory ×1 temporal evidence ×1 fast-weight memory ×1 Matrix Completion ×1 Axial Attention ×1 Synthetic Pretraining ×1 Cost-Latency Trade-off ×1 In-Parameter Memory ×1 Adapters ×1 Multi-device ×1 Action Guard ×1 Memory Provenance ×1 Hands ×1 Tool Revision ×1 Content Credibility ×1 Fake News ×1 Multi-branch Redundancy ×1 Path Reasoning ×1 Transformer Decoder ×1 Agentic Retriever ×1 Multi-step Search ×1 Recall@10 ×1 Open-weight ×1 voice assistant ×1 on-device LLM ×1 MRoPE ×1 SGLang Serving ×1 Stateless Agents ×1 sparse routing ×1 entmax ×1 sparsemax ×1 normmax ×1 Loop-aware Training ×1 Confidence Gate ×1 agent plasticity ×1 persistent artifacts ×1 held-out generalization ×1 learning efficiency ×1 SimHash ×1 Cost-aware Search ×1 Long-Context LLM ×1 Evidence Discovery ×1 Agentic policy discovery ×1 Amortized search ×1 Skeleton-based Prompting ×1 Language Selection ×1 Conversational Recommender ×1 Behavioral Replay ×1 Action Set Pruning ×1 Compositional Skill ×1 Cost Amortization ×1 CMP ×1 Workspace Interaction ×1 Task Tracking ×1 Environment-Side Memory ×1 Intrinsic Metrics ×1 LLM Mutation ×1 parallelization ×1 scheduling ×1 Contrastive learning ×1 Late interaction ×1 Evidence selection ×1 Harness Reliability ×1 Browser Automation ×1 Oracle ×1 Operating System ×1 Abstract Machine ×1 prospective learning ×1 Drafting ×1 LEGO ×1 Streaming Memory ×1 Retrieval Credit Assignment ×1 Evidence Coverage ×1 Retrieval Unit ×1 Hyperedge ×1 EDA ×1 prompting ×1 page layout ×1 offline evaluation ×1 secondary stack ×1 Mental Models ×1 Belief-Conditioned Reward ×1 Experience Retrieval ×1 Knowledge-Intensive Tasks ×1 Artifact Grounding ×1 Agentic BBO ×1 Black-Box Optimization ×1 Finite-Budget ×1 Relevance Assessment ×1 Backbone Evolution ×1 Prompt Robustness ×1 IR Evaluation ×1 N:M sparsity ×1 SpMV ×1 memory-bound ×1 Explainable IR ×1 Ideal Query ×1 Separability ×1 Post-hoc Analysis ×1 Text Games ×1 two-level softmax ×1 sampling bias ×1 candidate sampling ×1 3D Privilege ×1 Rulebook ×1 Code-as-Model ×1 Primer ×1 Multi-role LLM ×1 Parameter-free ×1 RL Scaling ×1 Groupwise Grading ×1 Training Infrastructure ×1 Agent Critic ×1 Feedback Tracking ×1 soft memory tokens ×1 LLM memory ×1 residual connection ×1 candidate completion ×1 listwise learning ×1 probability competition ×1 thinking mode ×1 no-thinking ×1 Skill Verification ×1 Scenario Synthesis ×1 Skill Library Curation ×1 Synthetic Environment ×1 GUI Benchmark ×1 Visual Pointing ×1 Test Evaluation ×1 Peer Cross Validation ×1 Token-Level Routing ×1 Delayed Batching ×1 Request-Centric Programming ×1 Model-Centric Execution ×1 Document Structure ×1 Verifiable Supervision ×1 Event Sequence Forecasting ×1 Generative Modeling ×1 Wasserstein Barycenter ×1 Accuracy-Realism Trade-off ×1 Agentic Continual Learning ×1 Reflection Revision ×1 Online Streaming ×1 Textual Feedback ×1 Long-horizon reasoning ×1 Linear-time inference ×1 Memory anchors ×1 Recurrent architecture ×1 O(1) memory ×1 latent recurrence ×1 elastic depth ×1 replay generation ×1 self-preservation ×1 Incremental Learning ×1 Structured Harness ×1 Open-Ended Generation ×1 Report Continuity ×1 Redundancy ×1 autoregression ×1 decoding schedule ×1 treedepth ×1 corruption lattice ×1 Epistemic Humility ×1 Evidence Routing ×1 Teacher-Relative Shift ×1 Distillation Geometry ×1 AgentSWE ×1 agent software engineering ×1 Adversarial Heuristic Learning ×1 chain-of-thought faithfulness ×1 counterfactual evaluation ×1 legal NLP ×1 harness-native training ×1 Egress Control ×1 Zero Trust ×1 probes ×1 deception detection ×1 sabotage ×1 FIBS ×1 white-box ×1 ZCPO ×1 short video verification ×1 evidential sufficiency ×1 multimodal evidence ×1 fact-checking ×1 LLM verifier ×1 Multi-Stream Attention ×1 Depth Attention ×1 Multi-vector Retrieval ×1 Geopolitical QA ×1 POI Reranking ×1 Geospatial ×1 Heterogeneous Fusion ×1 Pairwise Ranking ×1 open-world tasks ×1 Harness-Native Training ×1 Execution-Reward RL ×1 SQL Agents ×1 Small LMs ×1 offline rollouts ×1 modality preference ×1 text-image fusion ×1 optimizer-state quantization ×1 4-bit AdamW ×1 preconditioner space ×1 Hidden States ×1 Matrix Entropy ×1 instruction selection ×1 prior barrier ×1 data budget allocation ×1 lifelong editing ×1 rank adaptation ×1 memory efficient ×1 Automated Feature Engineering ×1 Blockwise Search ×1 Memory Formation ×1 Context Grounding ×1 LoCoMo ×1 SEISMIC ×1 Forward Index ×1 Model-Skill Co-Evolution ×1 relational retrieval ×1 native topology ×1 long-form generation ×1 NTCIR-19 ×1 Trajectory Monitoring ×1 Streaming Alignment ×1 Early Intervention ×1 Environment ×1 self-generated QA ×1 serving ×1 Prefetching ×1 pointwise ranking ×1 sovereign AI ×1 morphology ×1 Document Tree ×1 Structure Navigation ×1 code-as-policy ×1 hierarchical decomposition ×1 JEPA-Anything ×1 Training Data Generation ×1 Sign Language Translation ×1 Gloss-free ×1 Text Selection ×1 Candidate Differences ×1 Laplace Approximation ×1 MuP ×1 Posterior Transfer ×1 Scale Transfer ×1 Visual Editing ×1 LMM-as-a-judge ×1 Value cache ×1 Symmetry ×1 Visual-Native ×1 Skill Induction ×1 Human Demonstration ×1 Multimodal Dataset ×1 scientific agents ×1 neural operators ×1 decision attribution ×1 hyperparameter optimization ×1 PDE ×1 Uncertainty-Guided ×1 Progressive Contraction ×1 Reasoning Post-Training ×1 Monte Carlo Reasoning Trees ×1 alignment generalization ×1 value representations ×1 activation analysis ×1 model behavior ×1 Relative Policy Difference ×1 KL-Regularized Optimization ×1 autoregressive retrieval ×1 query feedback ×1 token-level retrieval ×1 sparse token matching ×1 query adapter ×1 Tool Agents ×1 Role Specialization ×1 MX-NVFP4 ×1 Calibration-free ×1 Dual-scale ×1 Go-Explore ×1 sandbox snapshots ×1 failure recovery ×1 runtime state ×1 Feature Deprecation ×1 Feature Importance ×1 Perturbation ×1 Subsequence Association ×1 RL credit assignment ×1 reward redistribution ×1 generative recommender ×1 delayed rewards ×1 conservative hierarchical operator ×1 Asynchronous Rollouts ×1 INT4-INT8 ×1 Self-Conditioning ×1 Spiking Neural Networks ×1 Physics-Informed Neural Networks ×1 Automated Tuning ×1 LLM Code Generation ×1 Reference Selection ×1 difficulty progression ×1 dynamic data selection ×1 residual representations ×1 ANN search ×1 embedding calibration ×1 semantic workspace ×1 evidence support ×1 generative workspace ×1 Group Policy Optimization ×1 Critic-free RL ×1 Index-side optimization ×1 Semantic Restructuring ×1 Metadata Synthesis ×1 Reliability-Aware ×1 linear response maps ×1 activation editing ×1 task context ×1 state cache ×1 randomized transport ×1 STORM ×1 masking ratio ×1 encoder ×1 Behavior Supervision ×1 Web Interface ×1 Web Navigation ×1 personal agents ×1 tool dependency graph ×1 execution feedback ×1 multi-service ×1 MLP Head ×1 No Fine-Tuning ×1 Latent Evaluation States ×1 Intrinsic Reward ×1 LLM Assistant ×1 policy learning ×1 Wikipedia Game ×1 critique ×1 error localization ×1 token-level credit ×1 Temporal Ordering ×1 Specialist ×1 Generalist ×1 Model Editing ×1 Long-context Serving ×1 Hierarchical Sparse Attention ×1 Load-Adaptive Residency ×1 DFlash2-DSpark ×1 Verifier Hardening ×1 Agent Environments ×1 Empirical Bayes ×1 Interface Design ×1 Benchmark Difficulty ×1 Runtime Failure Prediction ×1 Latent Augmentation ×1 LLM Auditor ×1 Representation Perturbation ×1 Multi-fidelity ×1 Code Optimization ×1 Orchestrator ×1 Symbolic User ×1 length extrapolation ×1 Embedding Perturbation ×1 Low-rank optimization ×1 Spectral structure ×1 Adaptive rank ×1 Heterogeneous Coevolution ×1 Reinforced Self-Training ×1 Harness Search ×1 Relevance Prediction ×1 Cowork agents ×1 Token-to-Go ×1 Disaggregated Training ×1 Edge-Cloud Collaboration ×1 RL Fine-Tuning ×1 SLM-LLM Collaboration ×1 Hopper Kernel ×1 Gradient Bottleneck ×1 Hierarchical Loss ×1 Attribute-aware ×1 Bit-Factorized ×1 Zero-shot Adaptation ×1 Multi-tree Beam Search ×1 Semantic Hashing ×1 Proof Verification ×1 Rubric Reward Model ×1 Tool-Augmented Judge ×1 Preference Accuracy ×1 Socratic Learning ×1 DHPO ×1 Virtual Head Attention ×1 Grouped-Query Attention ×1 Reparameterization ×1 PAC-Learning ×1 Software Scaffolding ×1 ActionOPD ×1 Semantic Actions ×1 DRAM-free ×1 SSD Index ×1 legal IR ×1 document segmentation ×1 metadata enrichment ×1 business benchmark ×1 execution validity ×1 policy compliance ×1 object-oriented context ×1 Critique-Guided Rewriting ×1 Teacher Recoverability ×1 DAGFormer ×1 cross-layer connections ×1 topology predictor ×1 self-supervision ×1 loop detection ×1 Fashion Retrieval ×1 Attribute-Level ×1 LLM Query Generation ×1 Temporal Memory ×1 Privileged Adaptation ×1 HybridLink ×1 completion graph ×1 rejection sampling fine-tuning ×1 agent-as-judge ×1 executable verification ×1 Decision Architecture ×1 Cacheable ×1 Null Detection ×1 Sequential Experimental Design ×1 Bayesian Experimental Design ×1 KSA ×1 C-KSA ×1 S-KSA ×1 Looped MoE ×1 Asymmetric Compute ×1 Latency-Accuracy Tradeoff ×1 Open-source Platform ×1 Tool Learning ×1 Bidirectional KL ×1 Teacher Guidance ×1 Process Verification ×1 Token-level Reward ×1 Reward Attribution ×1 Recurrent States ×1 Qwen3.8-27B ×1 DSpark ×1 Decoding Acceleration ×1 SERO ×1 oracle continuation ×1 policy-reachability weighting ×1 long-horizon RL ×1 Social Agents ×1 Outcome Verification ×1 Fine-grained Visual Understanding ×1 Generation-side Representation ×1 GAFU ×1 Profile Enrichment ×1 Skill Credit Assignment ×1 Policy-Adaptive ×1 Skill Injection ×1 Progress Reward ×1 Entmax ×1 Logit Adjustment ×1 Fenchel-Young Loss ×1 Tool Discovery ×1 Retrieval Architecture ×1 Plan-then-Retrieve ×1 ToolBench ×1 unembedding ×1 pretraining dynamics ×1 crosscoding ×1 readout ×1 Surprisal rewards ×1 Self-play ×1 Verifiable rewards ×1 Direct File Access ×1 Advertising Delivery ×1 AuctionNet ×1 Event Forecasting ×1 Brier Score ×1 Evidence Acquisition ×1 Selective Retrieval ×1 Contrastive Retriever ×1 AutoPostTrain ×1 Weight Merging ×1 SFT Data ×1 Data Action Space ×1 Automated Data Engineering ×1 Multi-Query Retrieval ×1 Mixed-Precision Quantization ×1 Contrast Calibration ×1 e-commerce image generation ×1 multi-condition generation ×1 VLM agent ×1 FFN Update ×1 Spatiotemporal ×1 remote tier ×1 context parallelism ×1 LMCache ×1 RL-zero ×1 Multi-Session ×1 MDA ×1 routed-SDA ×1 agent routing ×1 specialization ×1 incentive design ×1 Forward-Deployed Engineer ×1 Customer Delivery ×1 Business Metric ×1 weak-strong collaboration ×1 semantic verification ×1 Benchmark Overfitting ×1 token credit ×1 single rollout ×1 Autoregressive Masked Language Modeling ×1 Multimodal Decision ×1 cascade delegation ×1 JEV ×1 cost-latency optimization ×1 Evidence-blind Supervision ×1 Crosscoder ×1 Attribute-Object Binding ×1 MLLM Embedder ×1 Synthetic Text Supervision ×1 Workflow Normalization ×1 weak-to-strong escalation ×1 Mahalanobis distance ×1 black-box oversight ×1 code execution ×1 harness-to-model decision transfer ×1 OSWorld-v2 ×1 real-device harness ×1 agent efficiency ×1 Trajectory Forecasting ×1 test-time training ×1 attention distillation ×1 Latent Compaction ×1 Graph Collaborative Filtering ×1 Dual-Branch ×1 Self-Orchestration ×1 Market-Based Bidding ×1 Decentralized Swarm ×1 Cross-evidence Relations ×1 Knowledge-state Updating ×1 CLI Agent ×1 Steerability ×1 Task-State Governance ×1 Recoverable Execution ×1 Interaction Paradigm ×1 Workflow Orchestration ×1 Reasoning Exploration ×1 forget gate ×1 selective erasure ×1 Orthogonal Transform ×1 Spectrum Alignment ×1 Butterfly ×1 Emotion-Conditioned Interaction ×1 Interaction-Aware Training ×1 reward shaping ×1 Agentic Deployment ×1 Self-improving ×1 Sequence Transduction ×1 Behavior Memory ×1 Disaggregated Serving ×1 Checkpoint ×1 RL distillation ×1 policy shift ×1 gated transfer ×1 Auxiliary Targets ×1 Action Forecasting ×1 First-Price Auction ×1 Price of Anarchy ×1 Agentic Market ×1 Mechanism Design ×1 3D coding ×1 Automated Verification ×1 Activation Compression ×1 Sparse ReLU ×1 VRAM ×1 Lossless Compression ×1 Memory writer ×1 Intrinsic Evaluation ×1 Fact Evolution ×1 Residual Stream ×1 sample-wise noise ×1 pragmatics ×1 social context ×1 Belief Collapse ×1 Conformity ×1 Diversity Preservation ×1 test-time computation ×1 context optimization ×1 logits steering ×1 LLM-as-Researcher ×1 Cost-Efficient Measurement ×1 weak-to-strong routing ×1 disagreement signal ×1 goal-adaptive ×1 agent traces ×1 Environment-Agnostic ×1 Control Flow ×1 Hierarchical Task Network ×1 Result-Informed Planning ×1 random projection ×1 data extraction ×1 Search-Augmented LLM ×1 Evidence Use ×1 Knowledge State ×1 Few-Shot Prompting ×1 Infilling ×1 Likelihood-Ratio Reward ×1 Text Editing ×1 LLM honesty ×1 utility preservation ×1 Recurrent Transformer ×1 Input Injection ×1 Architecture Design ×1 DLRM ×1 GEMM ×1 Hardware Co-design ×1 Posterior Distillation ×1 Off-Policy Correction ×1 Subspace Tuning ×1 LangDiT ×1 Generation Perplexity ×1 Training-Free Distillation ×1 Least Squares ×1 Agentic Model ×1 WIRL ×1 lookup table ×1 conditional memory ×1 Professional Work ×1 Information Acquisition ×1 Feasibility ×1 Unified Action Space ×1 Step-balanced GSPO ×1 Evaluation Awareness ×1 Refusal Behavior ×1 StreamingVQA ×1 ProactiveAgent ×1 CAUC ×1 Controllable Multi-Objective Recommendation ×1 PRISMA ×1 Systematic Review ×1 Data Extraction ×1 Project Memory ×1 Complementary Representations ×1 TSLM ×1 Agent Discovery ×1 Agent Card ×1 Candidate Context Scaling ×1 Training-free Retrieval ×1 Trajectory Conditioning ×1 Contextual Prior ×1 Librarian Agents ×1 Workspace Curation ×1 linear probing ×1 knowledge representation ×1 goal drift ×1 steering agent ×1 fast-slow paths ×1 Utility Estimation ×1 Set-level RL ×1 Diverse Generation ×1 Causal Retrieval ×1 routing exploration ×1 shared experts ×1 Neural Sparse Retrieval ×1 Harness-Policy Optimization ×1 Fine-grained captioning ×1 VLM training data ×1 Adaptive zooming ×1 FactCycle ×1 Visual information transfer ×1 Trace Construction ×1 Student-Conditioned ×1 Memory Gating ×1 Search Engine ×1 Pass@1 ×1 Experience Consolidation ×1 Causal Validation ×1 Memory Externalization ×1 Conditional Skill Induction ×1 Tool Recommendation ×1 CUDA kernel ×1 Transformer inference ×1 Hopper ×1 MCP gateway ×1 Hierarchical Architecture ×1 Field-scoped Behavior ×1 Feedback Evolution ×1 Cognitive Differentiation ×1 Cross-Source Reasoning ×1 Small Model ×1 Multi-agent Systems ×1 Agent Prioritization ×1 Black-box Optimization ×1 Memory-Aware Routing ×1 Session-Level LLM ×1 Cost-Quality Trade-off ×1 lossy verification ×1 Open-weight LLM ×1 Self-hosted ×1 Evaluation Suite ×1 Agent Personalization ×1 Dueling Bandit ×1 Information-Directed Sampling ×1 Dual-Channel Retrieval ×1 Time-aware RAG ×1 Selective Skill Exposure ×1 Reasoning Reranker ×1 OpenOneRec ×1 Progress Stagnation ×1 CodeAct ×1 model-harness co-evolution ×1 meta-agent ×1 rejection sampling SFT ×1 brain-body mismatch ×1 Cowork Agent ×1 Group-Relative ×1 Error Reproduction ×1 Abstraction ×1 Cost-Performance Trade-off ×1 LLM-Generated Tests ×1 training harness ×1 JIT critic ×1 Ranked Preference ×1 f-divergence ×1 agentic reranking ×1 reasoning-intensive IR ×1 hard negative mining ×1 Retail QA ×1 Schema Briefing ×1 Cloud Object Storage ×1 RaBitQ ×1 Request Merging ×1 Mode Selection ×1 Multi-Model Coordination ×1 Control Allocation ×1 Shared Autonomy ×1 Prefill Efficiency ×1 verifier generation ×1 executable verifiers ×1 Speculative Reasoning ×1 Realtime ASR ×1 Concurrent LLM ×1 Sales Agent ×1 Commercial Conversations ×1 implicit budgeting ×1 task composition ×1 reasoning tokens ×1 Semantic Association ×1 Amazon ×1 OpenDS-World ×1 Lifelong Agent ×1 Memory Worth ×1 RFM ×1 Feature Injection ×1 Distributional Supervision ×1 Dynamic Update ×1 Semantic Drift ×1 Runtime Monitor ×1 flow language model ×1 Decision model ×1 Multi-agent pool ×1 Knowledge graph ×1 Luce model ×1 schedule invariance ×1 asynchronous execution ×1 evidence acquisition ×1 CEA ×1 retrieval stability ×1 multi-vector retrieval ×1 attention pooling ×1 self-ensemble distillation ×1 dropout ×1 single-tower ×1 closed-form variance ×1 Habit Induction ×1 Missing Modalities ×1 Truncated Backpropagation ×1 Diffusion LMs ×1 Decoding Speed ×1 Cost Analysis ×1 Video Advertising ×1 Memory-augmented LLM ×1 Gated DeltaNet-2 ×1 Mixture-of-Storage ×1 Stateful Inference ×1 Cross-Layer Sharing ×1 Expert Redundancy ×1 Residual Pathways ×1 Neighborhood-Aware ×1 Spam Removal ×1 Ranking Consistency ×1 One-Pass Inference ×1 Fusion ×1 Search as Code ×1 Python Library ×1 Codex ×1 Decision Certificate ×1 Rationale Supervision ×1 looped language models ×1 recurrent depth ×1 norm anchoring ×1 cross-tokenizer ×1 closed API teacher ×1 Audio Understanding ×1 Low-latency Streaming ×1 Instruction-controlled ×1 Agentic Orchestration ×1 Self-Compaction ×1 Approximate Vector Search ×1 Large-K ×1 Compressed Scanning ×1 GUI-MCP-CLI ×1 Open-Source Models ×1 Quality Gates ×1 ML for Science ×1 Evaluation Gaming ×1 Scientific Hill-Climbing ×1 Sim2Real ×1 Iterative RAG ×1 Knowledge cutoff ×1 Multimodal Reranking ×1 System One Decision ×1 Zero-shot Reranking ×1 Decision Tuning ×1 long-sequence modeling ×1 two-stage retrieval ×1 user behavior sequence ×1 industrial recommendation ×1 online experiment ×1 baseline-incremental ×1 Search-State Awareness ×1 LASAGNA ×1 Interface Evolution ×1 Population-based Evolution ×1 Proxy Evaluation ×1 LLM Mutator ×1 Agent Steering ×1 Pitfall Anticipation ×1 Latent World Model ×1 MPC ×1 Language Actions ×1 Policy Fine-tuning ×1 Object-Oriented Design ×1 AutomationBench ×1 Orthogonal Design ×1 Cross-Harness Transfer ×1 serving system ×1 position encoding ×1 context engineering ×1 Difficulty Measure ×1 Coverage Gap ×1 Measure Mismatch ×1 Item Rewriting ×1 Personal Assistants ×1 Prefix Balancing ×1 Personalized Decoding ×1 3DGS ×1 Elo Tournament ×1 Continuous Tokens ×1 style transfer ×1 CycleGRPO ×1 Temporal Grounding ×1 agentic forecasting ×1 skill distillation ×1 environment evolution ×1 Recursive Block ×1 LoopRec ×1 Attribution Gap ×1 Production Metrics ×1 Lineage Schema ×1 Latent Interconnect Protocol ×1 connector training ×1 representation exchange ×1 Temporal Sensitivity ×1 Generative Relevance Feedback ×1 Hierarchical LM ×1 Continuous Latent ×1 Multimodal CTR ×1 Progressive Learning ×1 Query-Item Relevance ×1 Creative Discrimination ×1 Graph-based ANN ×1 Data Weighting ×1 Skill Withdrawal ×1 Capability Framework ×1 Off-Policy ×1 Fairness-aware Recommendation ×1 Agentic Policy Selection ×1 Exposure Fairness ×1 Verified Repair ×1 LLM Auditability ×1 Runtime Control ×1 soft prompt ×1 behavioral localization ×1 trajectory distillation ×1 skill adaptation ×1 GUI Interaction ×1 CLI Systems ×1 Kernelized Attention ×1 Nonstationary Kernel ×1 LLM classifiers ×1 cascading ×1 ensembling ×1 self-synthesis ×1 Search Momentum ×1 Query Intent Drift ×1 Rubric Tree ×1 Multimodal Evidence ×1 Active Observation ×1 Factuality Evaluation ×1 agentic commerce ×1 transaction authority ×1 authorization boundary ×1 representation distillation ×1 Viewpoint Coverage ×1 Marginal Utility ×1 Ranker ×1 Co-occurrence ×1 unmasking policy ×1 controlled study ×1 Controlled Evaluation ×1 temperature ×1 Dreaming Learning ×1 Per-Layer Embeddings ×1 Cross-Layer Latent ×1 OPTIC ×1 Lossy Decoding ×1 TWIG ×1 Weight Sharing ×1 Discrete Optimization ×1 AI-Native Engine ×1 Game Generation ×1 Schema-Validated State ×1 data voids ×1 web-searching LLMs ×1 disinformation ×1 object detection ×1 segmentation ×1 referring localization ×1 pixel decoder ×1 two-stage training ×1 PubMed ×1 Claim Verification ×1 Integrity Violations ×1 Runtime Intervention ×1 document ingestion ×1 formatting cues ×1 enterprise agents ×1 OOXML ×1 memory deletion ×1 post-deletion persistence ×1 Trajectory Visualization ×1 Decision Transparency ×1 Non-Stationary ×1 Shadow Replay ×1 Bézier Flow ×1 Residual Flow ×1 looped architectures ×1 KV-sharing ×1 stochastic depth ×1 skill granularity ×1 runtime routing ×1 macro-to-micro recovery ×1 finite-state executor ×1 Specification Engineering ×1 Efficient LLM ×1 causal masking ×1 future-token leakage ×1 teacher forcing ×1 model auditing ×1 prefix invariance ×1 n-gram embeddings ×1 low-bit ×1 memory optimization ×1 steering adapter ×1 controllable text generation ×1 Attention Optimization ×1 Audio-Visual Retrieval ×1 Structure-Grounded Reasoning ×1 Active Information Gain ×1 Multi-turn Inference ×1 Over-representation analysis ×1 Bioinformatics ×1 fusion tax ×1 Typed Feedback ×1 Bounded State ×1 revision marker ×1 Sequence Compression ×1 Visual Token Merging ×1 Test-time Regression ×1 tree attention ×1 parallel inference ×1 policy-exact ×1 Neuro-symbolic ×1 Sequential Rules ×1 Next-Event Prediction ×1 Residual Network ×1 Event Sequences ×1 performance prediction ×1 self-confidence ×1 cascades ×1 Semantic Search ×1 Mathlib ×1 state integrity collapse ×1 long-horizon dialogue ×1 behavioral drift ×1 structured state anchoring ×1 Certified Screening ×1 Long-Document VQA ×1 Adaptive Rubrics ×1 Neural Retrieval ×1 MICE ×1 Efficiency-Effectiveness ×1 CoT Distillation ×1 Foundation Recommendation Models ×1 preference learning ×1 history modeling ×1 people recommendation ×1 Framework ×1 depth growth ×1 induction heads ×1 parameter reuse ×1 trust calibration ×1 automation bias ×1 numerical stability ×1 grokking ×1 Shape-Gain ×1 Spherical Optimization ×1 Panorama Generation ×1 CubeAtlas ×1 Super-Resolution ×1 selective auditing ×1 online error control ×1 confidence scores ×1 Gumbel noise ×1 Long-Horizon Shopping ×1 Cross-domain Transfer ×1 Retriever-agnostic ×1 Attention Leakage ×1 metaphor ×1 causal subspace ×1 Decision Theory ×1 Risk-Cost ×1 Update Cost ×1 Retrieval Quality ×1 Wikipedia Revisions ×1 thought pruning ×1 quality-diversity tradeoff ×1 Agentic Auditing ×1 Web Corpus ×1 FineWeb2 ×1 LLM Debugging ×1 Executable Tests ×1 spatiotemporal ×1 Elastic Inference ×1 graph diffusion ×1 memory-augmented LLM agents ×1 triple retrieval ×1 GTV Estimation ×1 Set-Aware Encoder ×1 O2O Advertising ×1 Residual Ensemble ×1 precision objective ×1 FORK ×1 Discriminative-Generative Fusion ×1 CRM ×1 Stateful Environments ×1 Harness Evaluation ×1 Model-Harness Fit ×1 Learning frontier ×1 Teacher-student ×1 Multi-turn agents ×1 Query Robustness ×1 Syntactic Variation ×1 Writing Errors ×1 Self-Improving ×1 Research Automation ×1 Decision Reuse ×1 Jacobian ×1 TopKLoRA ×1 Sparse Fine-tuning ×1 Circuit Discovery ×1 Backdoor ×1 Cross-space Distillation ×1 information extraction ×1 Metadata Registry ×1 Artifact Provenance ×1 Patent Examination ×1 Prior-Art Retrieval ×1 Input Preprocessing ×1 Visual Cues ×1 Experience Synthesis ×1 Trajectory Aggregation ×1 Inference Cost ×1 Numerical Stability ×1 agent skill retrieval ×1 background projection ×1 embedding anisotropy ×1 IR ×1 Decision Grounding ×1 Compositional Coverage ×1 KFAC-SVD ×1 Layer-wise Error ×1 agentic memory ×1 information retrieval ×1 Video Advertisement ×1 Shot Selection ×1 Computational Advertising ×1 Revision ×1 Tool Ecosystem ×1 Capability Evolution ×1 Supervised Fine-tuning ×1 Cross-stage Information Flow ×1 Neuro-Symbolic Programs ×1 Agent Workflows ×1 Recurring Tasks ×1 LLM Calls ×1 GGUF ×1 model registry ×1 acceptance testing ×1 Memory Tax ×1 advantage shaping ×1 Interface Representation ×1 DART ×1 naive laddering ×1 credence good ×1 Coordinator-Executor ×1 Shared Prompt ×1 Harness-Model Collaboration ×1 Token-level Signal ×1 Multimodal Coding Agents ×1 Reference-free ×1 Code Editing ×1 Agent Capabilities ×1 outcome RL ×1 pass@1 ×1 finishing ×1 reach ×1 Capability Tree ×1 Real2Sim2Real ×1 Agent Router ×1 decision utility ×1 search gain ×1 Native Skill Discovery ×1 Commit-Order Prediction ×1 SDAR ×1 Scalar 2-bit ×1 MATH-500 ×1 financial-reasoning ×1 capacity ×1 load ×1 GPT-4o mini ×1 statistical modeling ×1 proactive memory ×1 false alarm control ×1 multiple testing ×1 Context-Aware Reranker ×1 KL-Regularized Control ×1 Bundle Construction ×1 Relay Thinking ×1 Confidence-guided ×1 MACR ×1 Multi-reward RL ×1 Pareto Dominance ×1 Hallucination Reward ×1 Attention Head Norms ×1 Multi-Scenario ×1 item memory ×1 personalized retrieval ×1 caching ×1 Cost-Efficient ×1 Recommendation Explanation ×1 Monte Carlo Sampling ×1 Chain of Debate ×1 cross-day compatibility ×1 scale management ×1 orthogonal alignment ×1 Deep CTR Prediction ×1 Input-side Selection ×1 Embedding Pruning ×1 Sample Reweighting ×1 MLE agent ×1 confidence gating ×1 execution prediction ×1 Channel Selection ×1 Ternary Sketch ×1 Low-rank Recovery ×1 Navigational Search ×1 Web-scale Retrieval ×1 persona prompting ×1 synthetic query generation ×1 semantic parsing ×1 LLM hallucination ×1 List-wise Generation ×1 Single Greedy Decoding ×1 Agentic Serving ×1 Tool-Aware ×1 Automatic Evaluation ×1 Information Planning ×1 Repair Loop ×1 WorkPilot ×1 Item-item similarity ×1 Phantom correlation ×1 User activity ×1 Rasch model ×1 Popularity bias ×1 Implicit feedback ×1 diffusion VLM ×1 visual routing ×1 certificate routing ×1 Adaptive Curriculum ×1 Zero-Data ×1 Proactive Interaction ×1 Ask-F1 ×1 Idea Injection ×1 Explainable Retrieval ×1 multi-hypothesis learning ×1 winner-takes-all ×1 multi-interest ×1 annealed MCL ×1 head collapse ×1 Task State ×1 Multi-State Linear Attention ×1 Second-Order Routing ×1 Mamba-2 ×1 Diffusion Recommender ×1 Generative Diversity ×1 Posterior-Predictive Aggregation ×1 Diversity-Accuracy Trade-off ×1 Straight-Through Gradient Routing ×1 Multi-Scenario Recommendation ×1 CP Decomposition ×1 multimodal embedding ×1 Multi-view Retrieval ×1 Affect-aware Memory ×1 Entropy Routing ×1 data filtering ×1 Reflective Learning ×1 graph-neural-network ×1 linear-time ×1 hierarchical-modeling ×1 syntax-free ×1 Environment-Aware ×1 Batch Selection ×1 Communication ×1 Trading agents ×1 Herding ×1 Data Composition ×1 Multi-Objective Distillation ×1 Uplift Modelling ×1 Promotion Allocation ×1 Monotonic Response ×1 Large Commerce Model ×1 Level-Specific Attention ×1 Encoder Decoder ×1 Manifold Learning ×1 Soft Conceptor ×1 Heterophily ×1 graph compression ×1 Stream VByte ×1 weight-norm control ×1 Gram-Schmidt ×1 question-evidence supervision ×1 submodular optimization ×1 Spectral Graph ×1 Multi-Hop Retrieval ×1 Catalog Anchoring ×1 Pareto Selection ×1 team decision-making ×1 communication gap ×1 retrieval gap ×1 pivotal information ×1 Evidence Units ×1 HTML Structure ×1 Budget-Neutral ×1 Investment Research ×1 Narrative-Driven Recommendation ×1 Preference Dependency Graph ×1 Open-Web Retrieval ×1 Topological Ranking ×1 Teacher Evolution ×1 temporal grouping ×1 behavior sequence modeling ×1 policy collapse ×1 self-reinforcement ×1 collective dynamics ×1 majority vote ×1 query batching ×1 disk-resident graph ×1 cooperative search ×1 I-O optimization ×1 Shared Prefill ×1 Over-proactivity ×1 Timing-Intensity ×1 Reasoning Effort ×1 Ranking Aggregation ×1 Quantum-Inspired ×1 Active Feedback ×1 asynchronous sampling ×1 Computer Use Agent ×1 Hybrid GUI-Code ×1 Synthetic QA ×1 distribution matching distillation ×1 EMA features ×1 weight normalization ×1 effective learning rate ×1 Contextual Embeddings ×1 Multi-vector ×1 Specification Synthesis ×1 Long-Horizon Coding ×1 Unified Training ×1 RL as Likelihood ×1 Platonic Representation ×1 multimodal alignment ×1 image-text alignment ×1 Multimodal Parsing ×1 Unified Schema ×1 Self-Evolving Training ×1 Agent Perception ×1 citation prediction ×1 reranker distillation ×1 GiGPPO ×1 Verification Record ×1 Listwise ×1 personalized learning ×1 adaptive learning ×1 Multi-episode RL ×1 Advertiser Subsidy Allocation ×1 Capability Licenses ×1 EPCG ×1 Relational Memory ×1 Omnimodal Embeddings ×1 Joint Embedding ×1 Embedding Unification ×1 Typed Plan ×1 Deterministic Operators ×1 Training Utility ×1 NDCG@10 ×1 Gated Linear Attention ×1 Initialization ×1 RULER ×1 Linear Recurrence ×1 question decomposition ×1 QA caching ×1 self-contained sub-questions ×1 Multi-Router ×1 LatentMoE ×1 Sparse Mixture-of-Experts ×1 Agentic IR ×1 dynamic environments ×1 stale information ×1 grading ×1 certification ×1 Team Organization ×1 Decision Graph ×1 Clinical Decision Support ×1 Coverage-aware retrieval ×1 Query decomposition ×1 Nugget evaluation ×1 Self-feedback ×1 Long-form QA ×1 Alias ×1 Auxiliary Rewards ×1 Offline Weighted SFT ×1 Small LM ×1 Gist Ranking ×1 Robotics Simulation ×1 RL Environments ×1 Web Accessibility ×1 Interactive Tools ×1 WCAG ×1 Subgraph Retrieval ×1 Memory Indexing ×1 Value-Indexed Memory ×1 Long-Context Agents ×1 Traversal ×1 Training-Free Adaptation ×1 Modular Knowledge ×1 Regression Testing ×1 Federated Recommendation ×1 Knowledge Sharing ×1 Recommender Agents ×1 Reasoning Integration ×1 On-policy Self-distillation ×1 Token-level Teacher Selection ×1 Multi-teacher ×1 molecular discovery ×1 workflow optimization ×1 cross-task memory ×1 generative user representation ×1 ads ranking ×1 exploration weighting ×1 LLM-guided search ×1 sequential decision ×1 high-recall IR ×1 Token-level KL ×1 prefix executability ×1 Retrieval Feedback ×1 exponential tilt ×1 inference compute ×1 Automated Design ×1 Coarse Routing ×1 Research Evaluation ×1 multilingual safety ×1 template-based ×1 low-resource languages ×1 multi-turn defense ×1 shallow alignment ×1 decision token ×1 statistical hypothesis testing ×1 clinical timelines ×1 LLM watermarking ×1 Gumbel-max ×1 Gaussian copula ×1 distortion-free ×1 Multi-Business Recommendation ×1 Permutation Preference Optimization ×1 Advertising Debiasing ×1 GSP Auction ×1 Computer-Using Agents ×1 Agentic Exploration ×1 GUI Checks ×1 Code Checks ×1 layer-wise allocation ×1 Density Ratio Estimation ×1 Direct Method ×1 Counterfactual Replay ×1 Mixture Weighting ×1 Logit Perturbation ×1 Inference-Time Ensemble ×1 Neural Thickets ×1 PSV ×1 LLM Code Evolution ×1 Early Ranking Feedback ×1 Unsupervised Pretraining ×1 Cross-Passage ×1 expert pruning ×1 router semantics ×1 replacement operator ×1 Zero-label ×1 Tool Efficiency ×1 Clinical Genomics ×1 Task Retrieval ×1 Dependency Orchestration ×1 Personalized Embeddings ×1 Decoding Length Prediction ×1 Prefill ×1 Stochastic Process ×1 Central Limit Theorem ×1 Long-sequence Recommendation ×1 Latent Workspace Ranker ×1 Evidence Scaling ×1 Reasoning Scaling ×1 Sparse Events ×1 Distributional Evaluation ×1 TaobaoBrowse ×1 time series foundation model ×1 multivariate forecasting ×1 covariates ×1 Chronos-2 ×1 Cross-Domain POI Recommendation ×1 Mobility Behavior ×1 CoT Causality ×1 Reasoning Faithfulness ×1 decision model ×1 first-token ranking ×1 RankNet ×1 context prefill ×1 Prefix Sharing ×1 State Forking ×1 clarification requests ×1 shared attention ×1 approximate DP ×1 Agent Interface ×1 Accessibility ×1 Programmatic ×1 Identity Behavior Modeling ×1 Transaction Risk ×1 High-Cardinality ID ×1 Sequence Foundation Model ×1 Hierarchical Indexer ×1 Subquadratic ×1 robust-preference-learning ×1 noise-correction ×1 LLM-fine-tuning ×1 Decision Procedure ×1 Vector-free ×1 multi-tool completeness ×1 retrieval structures ×1 cost analysis ×1 memory-efficient optimizer ×1 momentum-free ×1 Latent Sparse Attention ×1 Continuous Diffusion Models ×1 Long Text Generation ×1 SecPO ×1 Knowledge Compilation ×1 Claim-Evidence ×1 RAGFlow ×1 TSBPO ×1 tree-structured sampling ×1 language agents ×1 Multi-Agent Negotiation ×1 Execution-Grounded ×1 Adaptive Strategy ×1 Flow Map ×1 One-step Generation ×1 stepwise querying ×1 CoT-free ×1 SEIG ×1 Experience Flywheel ×1 Staged Learning ×1 Bayesian Bootstrap ×1 trajectory collection ×1 history-guided ×1 train-test mismatch ×1 Temperature Invariant ×1 TKGQA ×1 Agentic Recommendation ×1 Dynamic Reflection Tree ×1 Self-Evolving Recommender ×1 Runtime Guardrails ×1 Knowledge Repository ×1 Good Abandonment ×1 Search Metrics ×1 Randomized Experiments ×1 User Success ×1 Systems Optimization ×1 factual recoverability ×1 Evidence-Guided ×1 VectorDB Metadata ×1 Trajectory Reconstruction ×1 E-Commerce Agent ×1 Evaluation Chain ×1 Page-Native Retrieval ×1 Counterfactual Credit Reweighting ×1 Refutation ×1 Pre-execution Verification ×1 quality-cost trade-off ×1 learned router ×1 general-purpose AI ×1 probabilistic scoring ×1 Multi-Teacher ×1 Self-Exploration ×1 Rank-Gain Utility ×1 ID-free ×1 Item Representation ×1 Automated Algorithm Discovery ×1 Mechanism Hypothesis ×1 Experiment Feedback ×1 Prompt Enhancement ×1 lookup-table ×1 block-scaled ×1 companding grid ×1 Inference Strategy ×1 Blueprint Library ×1 Multi-tier Routing ×1 Cross-task Transfer ×1 Adversarial GEO ×1 Brand Injection ×1 Tool Composition Graph ×1 SFT-RL ×1 Replay Verification ×1 Data Preprocessing ×1 Prompt Augmentation ×1 Utility-Aware Decoding ×1 Prefix Conditioning ×1 Sliding-Window Attention ×1 Parameter-Efficient LLM ×1 Agentic Communication ×1 Dual Profile ×1 Homophily ×1 Intent-aware ×1 Citation Rank ×1 semantic state ×1 selective reading ×1 agentic LLM ×1 Mix-Ranking ×1 data-constrained training ×1 compute efficiency ×1 Local Search ×1 Spatiotemporal Encoding ×1 Diversity Decoding ×1 Requirement Elicitation ×1 Tutorial ×1 GPU Serving ×1 Filtering Index ×1