mlsanityData to 14 Oct 2026

Papers

New on arXiv

Sun 27 Sep 2026

584 papers

  1. Showing versus Doing: Teaching by DemonstrationOffline RL and imitation learning from data → Robot learning from demonstration +1
  2. PreSO: Privileged Representation Isolation for On-Policy Self-DistillationKnowledge distillation → On-policy and self-distillation for LLMs
  3. VH-GPO: Verified Handoff for LLM Agent Reinforcement LearningLLM reasoning → Reinforcement learning for reasoning +1
  4. SharedSAE: a unified sparse dictionary across u-net blocks and denoising time for diffusion modelsDiffusion models → Diffusion foundations and sampling +1
  5. MoR-MLLM: Mixture of Recursions for Efficient Multimodal Large Language ModelsLLM reasoning → Multimodal reasoning +1
  6. Emo-Jev: Probabilistic Reasoning for Emotion Classification with JevLLM reasoning → Audio-language model reasoning
  7. BridgeCast: Bridging Ocean Wave Forecasts to Reanalysis via Flow Matching with Exogenous VariablesDiffusion models → Flow matching, time series and science
  8. Generalizable single-cell perturbation response prediction using energy-guided flow matchingDiffusion models → Flow matching, time series and science
  9. Conformal Factuality Control for Multi-Hop Retrieval-Augmented GenerationRetrieval-augmented generation → RAG robustness and adaptive retrieval
  10. FORUM: Frozen Outputs Reconciled Using Model Agreement for Visual GroundingMultimodal LLMs / vision-language models → MLLM training and visual instruction tuning
  11. Relevance Is Not Sufficient Evidence: Detecting Evidence Gaps Before Generation in RAGRetrieval-augmented generation → RAG robustness and adaptive retrieval
  12. Representational Simplicity and Circuit Size Dissociate in a Threshold-Dependent Way: A Controlled Test via Adversarial TrainingMechanistic interpretability → Circuits and mechanistic analysis
  13. SINGED: Correct Outputs Do Not Certify Safe Execution in LLM AgentsLLM agents and tool use → Agent threats, guardrails and governance
  14. Agentic Commerce Bench: Measuring Fraud Detection for Agents That Spend MoneyLLM agents and tool use → Agent threats, guardrails and governance
  15. CipherGenome: Homomorphic Inference for Genomic Mixture-of-ExpertsMixture-of-experts → Efficient MoE training and inference systems
  16. GenomeOcean Anywhere: Private WebGPU Inference for Genome MoEsMixture-of-experts → Efficient MoE training and inference systems
  17. GenoTrace: Inheritable Watermarks for Genome Foundation Model DistillationLLM safety and security → Watermarking generated content and models
  18. UOPD: Uncertainty-Aware Intervention for On-Policy Distillation of Multi-Turn AgentsKnowledge distillation → On-policy and self-distillation for LLMs
  19. Faithful Activation Verbalization: Reducing Hallucinations in LLM Representation InterpretationLLM alignment / RLHF / preference optimisation → Preference alignment for recommenders and RAG
  20. Position Aware Layer Queries for Test Time Training in Vision Language ModelsMultimodal LLMs / vision-language models → Prompt learning and adaptation of CLIP
  21. Maat: Independent Deterministic Contract-Based Governance for Multi-Agent LLM WorkflowsLLM agents and tool use → Software engineering coding agents
  22. Fisher-Informed Recalibration for Feedback-Based On-Policy Self-Distillation of LLMsKnowledge distillation → On-policy and self-distillation for LLMs
  23. RICE-Alpha: Reliability-Informed Correction with Event Graphs for LLM-Agent Stock ForecastingLLM agents and tool use → Financial trading and forecasting agents
  24. MetaSampling: Making Frame Samplers Efficient for Long-Video Question AnsweringMultimodal LLMs / vision-language models → Video understanding with MLLMs
  25. RewardExplainer: Learning Reward Model Explanations from Counterfactual Preference FeedbackLLM alignment / RLHF / preference optimisation → Reward modeling and RLHF
  26. Opera: A Verbal Critic Framework for Long-horizon Coding AgentsLLM agents and tool use → Software engineering coding agents
  27. Test-Time Spatial Reasoning for Robot Manipulation Using Generative Real-to-SimLLM reasoning → Spatial and embodied reasoning
  28. DynGraphAgentBench: A Benchmark for Agentic Lifecycle Control in Dynamic Graph Anomaly DetectionLLM agents and tool use → Software engineering coding agents
  29. Beyond Solo and Consistency: Vindicating Multi-Agent Debate via Conditional Progressive PruningLLM reasoning → Reasoning for agents, planning and code +1
  30. On the Token Value Inequality in Efficient ReasoningLLM reasoning → Prompting and chain-of-thought reasoning
  31. EEG-Fusion: Failure-Informed Source-Free Expert Routing for Robust Motor Imagery EEG DecodingMixture-of-experts → Medical and biomedical MoE
  32. Designing Reliable LLM-as-a-Judge Measurement Systems for Multi-Turn Business AgentsLLM agents and tool use → Social simulation and multi-agent debate
  33. Simple Diffusion Language Models Are More Effective Few-Step Generators Than ReportedDiffusion models → Discrete and text diffusion
  34. ReSync: Re-Aligning the Two Clocks of Asynchronous World-Action ModelsWorld models → Video-generation world models and simulators
  35. Behavioral Monitoring of JEPA World Models with Jacobian CentroidsWorld models → Latent predictive world models
  36. HyperMCTS: Hypergraph-Augmented MCTS for Long-Horizon LLM AgentsLLM agents and tool use → Reinforcement learning for LLM agents
  37. When Consent Outlives Context: Residual Authority Replay in Long-Lived AgentsLLM agents and tool use → Agent threats, guardrails and governance
  38. Residual-Stream Burden Shapes Representation Learning in Diffusion TransformersDiffusion models → Diffusion foundations and sampling
  39. Where Activation Sparsity and KV-Cache Sparsity Cross in LLM DecodingEfficient LLM inference and serving → KV cache optimization
  40. Faster Block-Diffusion Serving with Distribution-Free Risk GuaranteesDiffusion models → Discrete and text diffusion
  41. Prospective Interpretation Risk: Principled Communication Control Between LLMsLLM agents and tool use → General LLM agent frameworks and benchmarks
  42. DexTaG: Tactile-as-Guidance in Reinforcement Learning for Dexterous ManipulationDeep reinforcement learning methods → RL for robot locomotion and manipulation
  43. Counterfactual Rollout Replay: Forkable Environments as Free Process Rewards for Software Engineering AgentsLLM agents and tool use → Software engineering coding agents +1
  44. When Successful Strategies Fail: Adaptation to Environmental Novelty in Terminal AgentsLLM agents and tool use → Agent skill learning and evolution
  45. R2 Flow: Recursive Self-Improvement via Recursive Skill EvolutionLLM agents and tool use → Agent skill learning and evolution +1
  46. Program-Verified Self-Evolution for Vision-Language ModelsMultimodal LLMs / vision-language models → VLM benchmarks and visual reasoning
  47. ReDrive: Shaping Representations with World Modeling for End-to-End DrivingWorld models → World models for autonomous driving +1
  48. QwenGyre: An Elastic Reinforcement Learning Framework for Training xLong-Horizon AgentsLLM agents and tool use → Reinforcement learning for LLM agents +1
  49. ViBR-WM: Visual Bayesian Regression for World ModelingWorld models → Latent predictive world models
  50. ChemOPD: Multi-Teacher On-Policy Distillation for Multi-Task Chemical ReasoningLLM reasoning → Prompting and chain-of-thought reasoning +1
  51. CLIMB-flow: Coupled Linear Inverse posterior sampling via Multiscale-Based flowDiffusion models → Inverse problems and image restoration
  52. Achieve What You Imagined: Learning to Align Actions with Visual PlansWorld models → Embodied and robot world models +1
  53. Vestrum: Improving Agent Harnesses by Adapting Their Verification, Structure and MemoryLLM agents and tool use → Software engineering coding agents
  54. Augmenting Visual Anomaly Detection with Automated InterpretabilityMechanistic interpretability → Interpreting vision and multimodal models
  55. Dual-Vocabulary Language Model for Cross-Tokenizer DistillationKnowledge distillation → Knowledge distillation for LLMs
  56. Identical Runs, Different Results: Benchmarking AI Coding Agents on Open-Weight ModelsLLM agents and tool use → Software engineering coding agents
  57. Controlling Speaking Rate in Autoregressive TTS via Activation SteeringMechanistic interpretability → Activation steering methods
  58. CodeActionBench: Evaluating Agentic Code-as-Policy for Embodied ManipulationLLM agents and tool use → Multimodal and embodied agents
  59. Diffusion Reward ModelsLLM alignment / RLHF / preference optimisation → Diffusion and multimodal model alignment
  60. Do We Really Need KL Divergence for On-Policy Distillation of Large Language Models?Knowledge distillation → On-policy and self-distillation for LLMs
  61. Is your uncertainty map wrong, or is its target? Exact diagnostics for the Tweedie diagonal, and a gradient-free alternativeDiffusion models → Medical imaging and segmentation
  62. Surprising Success, Repeated Failure: Entropy-Guided Credit Assignment for Exploration in LLM ReasoningLLM reasoning → Reinforcement learning for reasoning
  63. Selecting Diverse SFT Traces Improves Post-RL GeneralizationLLM reasoning → Reinforcement learning for reasoning +1
  64. Evidence-Inference Reconstruction: When The Evidence Is Recalled But The Reasoning Goes WrongLLM reasoning → Knowledge- and tool-augmented reasoning +1
  65. Tokens Change, Structure Endures: Spectral Watermarking for Generated SpeechLLM safety and security → Watermarking generated content and models
  66. DEALS: Decentralized Expertise-Aware Load Serving for Multi-Agent LLM SystemsLLM agents and tool use → General LLM agent frameworks and benchmarks
  67. Principal Steering Subspaces for Online Adaptation of Frozen Generative Robot PoliciesVision-language-action models → Reinforcement learning and post-training +2
  68. SecProbe: Adaptive Evaluation of Coding Agents on Cybersecurity VulnerabilitiesLLM agents and tool use → Agent threats, guardrails and governance
  69. EfficientAgent: What Makes KV Cache Offloading Work for Concurrent Agents?LLM agents and tool use → Software engineering coding agents
  70. Positions Are Not Facts: The Mismatch Between KV Caches and MemoryEfficient LLM inference and serving → KV cache optimization
  71. AnyStep-WAM: Budget-Aligned Distillation and Adaptive Inference for World Action ModelsWorld models → Embodied and robot world models
  72. DuraS2ST: Chain-of-Thought and Reinforcement Learning for Duration-Aligned Speech-to-Speech TranslationLLM reasoning → Audio-language model reasoning
  73. MomWorld: Momentum-Aware Latent World Model for Long-Horizon Autonomous DrivingWorld models → World models for autonomous driving +1
  74. Constrained Edit Fields for Training-Free Flow EditingDiffusion models → Image generation, text-to-image and editing
  75. ALDER: Discovering the Laws of a World by Acting in ItWorld models → World models for language agents
  76. A Statistical Perspective on Knowledge Distillation: Foundations, Classical Methods, and Large Language Model ExtensionsKnowledge distillation → Knowledge distillation for LLMs
  77. GeoShrink: Accelerating Diffusion Transformers with Two Lines of CodeDiffusion models → Flow matching, time series and science
  78. Self-Designed Evaluators and Warm Memory for Long-Horizon AgentsLLM agents and tool use → Agent memory systems
  79. BIRD: Distilling Decision Boundaries into Rationales for MLLM AdaptationMultimodal LLMs / vision-language models → MLLM training and visual instruction tuning
  80. DuoOPD: Learning from Joint Teacher-Student Outcomes for Multi-Task On-Policy DistillationKnowledge distillation → On-policy and self-distillation for LLMs
  81. Domain-Adaptive Dual-Gating Mixture of Experts for Generalizable Speech Deepfake DetectionMixture-of-experts → Multimodal MoE models
  82. Does Adversarial Training Improve Generalization in Multi-View VLAs? Revealing and Mitigating View CollapseMultimodal LLMs / vision-language models → Embodied and 3D vision-language models +1
  83. Prompt-Anchored Residual Adaptation for Biomedical Vision-Language ModelsMultimodal LLMs / vision-language models → Prompt learning and adaptation of CLIP
  84. One Latent, Many Tokens: Jointly Learning Compressed Embeddings for Efficient Language DiffusionDiffusion models → Discrete and text diffusion
  85. Seeing and Solving Are Not Enough for Vision-Language ModelsMultimodal LLMs / vision-language models → VLM benchmarks and visual reasoning
  86. When Do Agents Help? Embedding, LLM and Agentic Alignment of Classical Texts and Their TranslationsLLM agents and tool use → Web, search, and research agents
  87. SWE-Game: Can Coding Agents Build the Games We Want?LLM agents and tool use → Software engineering coding agents
  88. Auditing Agent Actions through Query-Conditioned AttributionLLM agents and tool use → Agent threats, guardrails and governance
  89. COGNIT-Guard: Calibrated Standalone Direct-Decision Guardrails with Heterogeneous CPU-NPU Confidence Cascading under Explicit Latency and False-Positive ConstraintsLLM safety and security → Safety alignment, refusal and guardrails
  90. Learning Multimodal Embeddings with Evidence-Aligned ReadoutMultimodal LLMs / vision-language models → Document understanding and multimodal RAG
  91. AgentBoundary: Counterfactual Evaluation of Safety in Tool-Using LLM AgentsLLM agents and tool use → Agent threats, guardrails and governance +2
  92. Towards Eliminating Catastrophic Forgetting in the Curriculum Learning of Math Reasoning TasksLLM reasoning → Mathematical reasoning
  93. Demonstration-Free Success-Probability Reward Learning for Generalist Robot PoliciesLLM alignment / RLHF / preference optimisation → Reward modeling and RLHF +1
  94. Tsubame: Tree Replay for Diffusion-Based Speculative DecodingEfficient LLM inference and serving → Speculative decoding
  95. InfraVLA: Extending Vision-Language-Action Navigation with Infrastructure CamerasVision-language-action models → Generalist VLA models and benchmarks
  96. SafeMol: Dual-Modality Safety Alignment for Molecular Multimodal ModelsLLM alignment / RLHF / preference optimisation → Safety alignment and jailbreak attacks +1
  97. Trajectory Unlearning on LLM-based AgentsLLM agents and tool use → Reinforcement learning for LLM agents +1
  98. Safety Reconstructed: Generative Modeling via Masked Diffusion Builds Strong Safety GuardrailsDiffusion models → Discrete and text diffusion +1
  99. Climbing the Hill: Prompt Injection Red-Teaming Against Frontier Models with Curriculum Reinforcement LearningLLM safety and security → Prompt injection and agent security +1
  100. StoryEngine: A State-Grounded Agentic Framework for Video StorytellingLLM agents and tool use → Multimodal and embodied agents
All 584
  1. ParaAgent: Reinforcing Parallel Acting in Open-World Tool EnvironmentsLLM agents and tool use → Reinforcement learning for LLM agents
  2. SpatialSpeak: QA-Native Reconstruction with Local and Global Context for Spatial Chain-of-Thought ReasoningLLM reasoning → Spatial and embodied reasoning +1
  3. EAT: Expert Account Tracker for Efficient MoE InferenceMixture-of-experts → Expert routing strategies
  4. Quizzing the Translation: A Prover-Grounded Evaluation Metric for NLrightarrowFOLLLM reasoning → Reasoning benchmarks and evaluation
  5. Beyond One-Step Accuracy: State-Affine Latent Transition for Reliable Visual PlanningWorld models → Latent predictive world models
  6. IVT-Guard: All-in-One Reasoning Model for AI-Generated Content DetectionLLM reasoning → Multimodal reasoning
  7. Fill2SR: Repurposing Inpainting Diffusion Transformers for Real-World Super-ResolutionDiffusion models → Inverse problems and image restoration
  8. ForeFly: A Dual-Horizon World Action Model for Aerial Vision-Language NavigationWorld models → World models for navigation
  9. SLIP-VLA: Single-Step Latent Imagination for Policy Learning in Vision-Language-Action ModelsVision-language-action models → World model-based VLA
  10. MA-JEPA: Joint-Embedding World Models for Multi-Agent Reinforcement LearningWorld models → Latent dynamics models and model-based RL
  11. OOD Generalization as a Bifurcation ProblemDiffusion models → Flow matching, time series and science
  12. Steer2Grasp: Inference-Time Embodiment-Aware Steering for Diverse Physically Feasible Grasp DiffusionDiffusion models → Diffusion and flow policies for robotics
  13. Reasoning on the Simplex: Geometric Fixed-Point ModelsLLM reasoning → Test-time compute and efficient thinking
  14. Does Execution Require Target KV Fidelity? A Mixed-Fidelity KV Runtime for LLM ServingEfficient LLM inference and serving → Serving systems and request scheduling
  15. In-Token Learning for High-Fidelity Image Restoration via Diffusion TransformersDiffusion models → Inverse problems and image restoration
  16. SceneScaffold: Active Scene-State Construction for Unified 3D Scene UnderstandingMultimodal LLMs / vision-language models → Embodied and 3D vision-language models
  17. Source Anchoring for Physical Consistency in Flow Matching ModelsDiffusion models → Flow matching, time series and science
  18. RelaxKV: Recomputation Guided by the Query with Sparse Context Attention for Efficient KV Cache ReuseRetrieval-augmented generation → Generation and context integration +1
  19. Hamiltonian JEPA: Action-Conditioned World Models with an Inherited Control StateWorld models → Latent predictive world models
  20. Chameleon: Dynamic Format Adapter for Efficient DiffusionDiffusion models → Diffusion foundations and sampling
  21. LLMs Trust Their Own: Identity-Dependent Conformity in Multi-Agent SystemsLLM agents and tool use → Social simulation and multi-agent debate
  22. Domain-Adapted Diffusion Models for Conditional Independence TestingDiffusion models → Diffusion foundations and sampling
  23. DISCO: Distributed Long Context Scaling with Grounding-Reasoning DisaggregationLLM reasoning → Test-time compute and efficient thinking
  24. LiveOption: Evaluating LLM Agents in Structured Option Trading with Nonlinear PayoffsLLM agents and tool use → Financial trading and forecasting agents
  25. VIDEAS: Distilling Explicit Action Semantics from Demonstration Videos for World Models via Prior-Guided SimulationWorld models → Embodied and robot world models
  26. SphMind: Towards Robust, Training-Free VLM-based Spatial Reasoning with a 360 CameraLLM reasoning → Spatial and embodied reasoning +1
  27. What Shared Prefixes Hide: Trajectory Dropout for On-Policy DistillationKnowledge distillation → On-policy and self-distillation for LLMs
  28. Concept Score Relearning: A Unified Cross-Architecture Attack on Concept ErasureDiffusion models → Robustness, privacy and adversarial aspects
  29. Elucidating the Design Space of Regression-based Diffusion Reinforcement LearningDeep reinforcement learning methods → RL with diffusion and flow models
  30. Raven: The Harness of Harnesses for Composable Agentic IntelligenceLLM agents and tool use → General LLM agent frameworks and benchmarks
  31. MoGround: Measuring and Mitigating Modality Distraction in Vision-Language ModelsMultimodal LLMs / vision-language models → VLM benchmarks and visual reasoning
  32. APEX: An Extensible Model for Agent-Assisted Production SchedulingLLM agents and tool use → Software engineering coding agents
  33. Graph-Guided Repository Environment ConstructionLLM agents and tool use → Software engineering coding agents
  34. TeacherGRPO: Closing the Capacity Gap in Reasoning Distillation via Teacher AlignmentKnowledge distillation → On-policy and self-distillation for LLMs
  35. TTRSD: Test-Time Reinforcement Learning with Self-Distillation for Vision-Language ModelsMultimodal LLMs / vision-language models → VLM benchmarks and visual reasoning +1
  36. Resolving State-Representation Mismatch: State-Space Visual Reasoning for Open-Loop VLA PlanningLLM reasoning → Spatial and embodied reasoning +1
  37. FoldAttention: Declared-Reference Softmax for Fast Decode and Deterministic BackwardEfficient LLM inference and serving → KV cache optimization
  38. Dense Is Not Enough: Hierarchical Supervision Allocation for Long-Horizon On-Policy DistillationKnowledge distillation → On-policy and self-distillation for LLMs
  39. StarBOA: Real-Time Mamba State-Space Unrolling for Sparse Radar Micro-Doppler in ISAC NetworksState-space sequence models (S4, Mamba) → Point clouds, detection and remote sensing
  40. DataMagic: Authoring Data Videos through Declarative Multi-Agent OrchestrationLLM agents and tool use → Multimodal and embodied agents
  41. Evaluating System One Models for Agent Security Decisions: Reliability, Calibration, and Selective AutomationLLM agents and tool use → Agent threats, guardrails and governance
  42. Groupwise Selective State-Space Filtering for Accurate and Streaming Action Boundary DetectionState-space sequence models (S4, Mamba) → Video, motion and generative Mamba models
  43. COEVO: Co-Evolving Context and Parameters for Recursive Self-ImprovementSynthetic data and self-improvement for LLMs → Self-rewarding alignment and self-training
  44. CoViST: Visual Token Compression via Composable StatesEfficient LLM inference and serving → Visual token pruning in multimodal models
  45. Beyond Timestamps: Decision-Aligned On-Policy Distillation for Long-Horizon AgentsKnowledge distillation → On-policy and self-distillation for LLMs
  46. From Position Risks to Block Survival: Faster Generation for Diffusion Language ModelsDiffusion models → Discrete and text diffusion
  47. OLED-MoE: Accelerating MoE-Based dLLM Inference via Inter-Iteration Locality-Aware Expert OffloadingMixture-of-experts → Efficient MoE training and inference systems
  48. PulseQuant: Propagation-Guided Subspace Correction for 4-Bit Video Diffusion TransformersDiffusion models → Video generation
  49. NLPG: Natural-Language Policy Gradients for Self-Evolving Language AgentsLLM agents and tool use → Reinforcement learning for LLM agents +2
  50. API Secrets Should Never Become Tokens in the LLM's Vocabulary: A Threat Analysis of API Credential Handling in LLM Agent Systems and an Empirical Evaluation of a Vault-Mediated Execution BoundaryLLM agents and tool use → Agent threats, guardrails and governance
  51. Unmask the State: When Does State Adaptation Matter for Masked Diffusion Language ModelsDiffusion models → Discrete and text diffusion
  52. MultiEcho: An Experimental Science of Learned WorldsWorld models → Latent predictive world models
  53. OPERA: A Unified Omnimodal Progressive Spatio-Temporal Reasoning Agent for Referring Video SegmentationLLM reasoning → Multimodal reasoning
  54. Safe Score Matching: Diffusion Policies with Hamilton-Jacobi Reachability for Online Safe Reinforcement LearningDiffusion models → Diffusion and flow policies for robotics +1
  55. Beyond Conservatism: Recoverability-Conditioned Exploration for Model-Based Imitation LearningWorld models → Latent dynamics models and model-based RL +1
  56. Does Learning to Predict the World Help Agents Act? Auditing World-Model Post-TrainingWorld models → Latent dynamics models and model-based RL
  57. When to Evict, Not What to Keep: Draft-Guided Eviction for Training-Free KV-Cache CompressionEfficient LLM inference and serving → KV cache optimization
  58. CertMark: Distortion-Free Multi-Bit Watermarking with Certified DecodingLLM safety and security → Watermarking generated content and models
  59. Agentic Multi-Turn Reasoning: A Fairness ApproachLLM reasoning → Reasoning for agents, planning and code +2
  60. Robust Hierarchical Structures for Agentic Document AnalysisLLM agents and tool use → Web, search, and research agents
  61. Relevance Does Not Imply Applicability: Experience Activation for Personal GUI AgentsLLM agents and tool use → GUI and mobile agents
  62. Hesitation-Aware On-Policy Distillation for Diffusion Language ModelsDiffusion models → Discrete and text diffusion
  63. AquaWAM: A Dynamics-aware World Action Model for Underwater Embodied AgentsWorld models → World models for navigation
  64. Direct Hidden-State Alignment: Mapping and Controlling Preference Expression in LLMsLLM alignment / RLHF / preference optimisation → Reward modeling and RLHF
  65. The Error You See Is Not the Error You Made: Progression-aware Reasoning Origin for Reasoning Error LocalizationLLM reasoning → Test-time compute and efficient thinking
  66. TraceDance: An Automated System for Building Agent Behavior Benchmarks from Real-World Agent Deployment TracesLLM agents and tool use → Software engineering coding agents
  67. The Statistical Benefits of Multiple Responses for Learning from DemonstrationsOffline RL and imitation learning from data → Imitation learning methods
  68. Calibration, Not Answer Selection: Distilling Internal Confidence in Reasoning ModelsLLM reasoning → Test-time compute and efficient thinking
  69. Learning to Sell: Reinforcement Learning for Strategic Large Language Model Agents in Multi-Product MarketsLLM agents and tool use → Financial trading and forecasting agents
  70. ChronoFlow: Hierarchical Flow Matching for Irregular Time Series GenerationDiffusion models → Flow matching, time series and science
  71. Q-WAM: 4-Bit Quantization of World Action Models with Action-Subspace ProtectionWorld models → Embodied and robot world models
  72. VehDyn: A Driving World Model Benchmark for Vehicle DynamicsWorld models → World models for autonomous driving +1
  73. GTRL: Grounding Divide-and-Conquer Value Learning with Temporal DifferencesOffline RL and imitation learning from data → Diffusion and flow policies for offline RL +1
  74. ActionGround: Training-Free Runtime Refinement of Frozen VLA PoliciesVision-language-action models → Reasoning and chain-of-thought in VLA
  75. VGGT-Diff: Visual Geometry Meets Diffusion for Sparse-View Novel View SynthesisDiffusion models → 3D, point cloud and scene generation
  76. Minimax-Optimality of Posterior Sampling for Reinforcement LearningDeep reinforcement learning methods → RL theory and sample complexity
  77. Acoustic Progress Propagation for Long-Horizon Speculative Decoding in ASREfficient LLM inference and serving → Speculative decoding
  78. ActiveMem: Dynamic Latent Memory Trees for Long-Horizon AgentsLLM agents and tool use → Agent memory systems
  79. CodeSkill: Latent Skill Abstraction for Long-Horizon Code AgentsLLM agents and tool use → Reinforcement learning for LLM agents
  80. Schrödinger--Föllmer Actor--Critic: Diffusion Policy Improvement with Finite-Sample AnalysisDiffusion models → Diffusion and flow policies for robotics +1
  81. PARSEE-VAD: Efficient Training-Free Online Video Anomaly Detection via Proposition-Aware Reasoning and Streaming Evidence EscalationLLM reasoning → Multimodal reasoning +1
  82. Beyond Memory Construction: Rethinking Memory Access for LLM-based Conversational AgentsLLM reasoning → Knowledge- and tool-augmented reasoning +1
  83. RMB: Reward Model Boosting Mitigates Reward HackingLLM alignment / RLHF / preference optimisation → Reward modeling and RLHF
  84. When Do Models Admit They Are Wrong? Failure Disclosure Is Unstable Under Reinforcement LearningDeep reinforcement learning methods → RL training of language and agent models
  85. Scope-WM: Scoped Computation for Efficient Visual World ModelsWorld models → Embodied and robot world models
  86. Background Gradients Shape Memorization in Flow MatchingDiffusion models → Image generation, text-to-image and editing
  87. WorldAgent: Verification-Guided Agentic Physical World ConstructionLLM agents and tool use → Multimodal and embodied agents
  88. Structured Residual Connectivity Matters for Diffusion TransformersDiffusion models → Inverse problems and image restoration
  89. ReAL: Accelerating Flow Matching through Segment Advancement with Shared LookaheadDiffusion models → Flow matching, time series and science
  90. Teach Yourself Where to Look: On-Policy Attention Self-Distillation for ReasoningLLM reasoning → Prompting and chain-of-thought reasoning +1
  91. TAO-DA: Towards Autonomous Operation--A Dual-Arm Vision-Language-Action Model for Coordinated ManipulationVision-language-action models → Generalist VLA models and benchmarks
  92. FocusDrive: Reasoning with Visual Focus for Autonomous DrivingEnd-to-end autonomous driving → VLMs for autonomous driving
  93. Offline Policy Evaluation via Mixed Bellman Residuals and Adaptive Critic RepresentationsDeep reinforcement learning methods → RL theory and sample complexity
  94. Identifying Temporal Features within Transcoders for Time Sensitive Factual RecallMechanistic interpretability → Knowledge and representations inside LLMs
  95. SeOPD: Self-Evolving LLMs via Online Policy Distillation from Self-Generated Chain-of-ThoughtLLM reasoning → Prompting and chain-of-thought reasoning +2
  96. DeltaWAM: Change-Centric Visual Foresight via Delta Tokens for an Efficient World-Action ModelWorld models → Embodied and robot world models
  97. Dynamic Manipulation with World-Action Models via Counterfactual PlanningWorld models → Embodied and robot world models
  98. Perturb-and-Solve: Efficient Learned-Operator Conditioning for Latent Diffusion Inverse ProblemsDiffusion models → Inverse problems and image restoration
  99. Downstream-Aware Context Selection for Online In-Context Reinforcement LearningLLM agents and tool use → Reinforcement learning for LLM agents +1
  100. Notes on Generative Modeling for Feedback Control and PlanningDiffusion models → Diffusion and flow policies for robotics
  101. FOCUS: Benchmarking Retinal Model Generalization from Foundation Vision Encoders to Multimodal LLMsMultimodal LLMs / vision-language models → Medical vision-language models
  102. Where Do Test-Time Scaling and Training Fall Short in Individual Stance Prediction?LLM reasoning → Test-time compute and efficient thinking
  103. What Does a Skill Actually Do? Estimands and Evaluation Validity for Tool and Skill Use in LLM Agents: A Critical ReviewLLM agents and tool use → Agent skill learning and evolution
  104. Not Too Hard, Not Too Easy: Learning from Intermediate States for LLM Structured ReasoningLLM reasoning → Test-time compute and efficient thinking
  105. DroneWAM: Efficient World Action Model for Drone Visual NavigationWorld models → World models for navigation
  106. LiteEvo: Automated, Cost-Efficient Harness Evolution for Generalization to Unseen TasksLLM agents and tool use → Software engineering coding agents
  107. Beyond State-as-Action: Exploiting Command-State Discrepancy for Robot Imitation LearningOffline RL and imitation learning from data → Robot manipulation imitation learning
  108. Beyond the Training Horizon: Mechanisms and Limits of Length Generalization in Looped TransformersLLM reasoning → Test-time compute and efficient thinking
  109. On Device Agentic Operation Caches -- Classifier-Centric NL-to-Action GenerationLLM agents and tool use → General LLM agent frameworks and benchmarks
  110. Leaky Students: Membership Inference against On-Policy DistillationLLM safety and security → Membership inference attacks +1
  111. Flow-Matching-Based Protein Structure Tokenizer Made Efficient and EasyDiffusion models → Flow matching, time series and science
  112. Policy Plasticity Matters in Offline-to-Online Reinforcement Learning: Refitting Offline Policies for Online AdaptationOffline RL and imitation learning from data → Offline RL algorithms
  113. Train Together or Merge Later? Unifying VLA Experts via a Shared Action InterfaceVision-language-action models → VLA architectures and generalization
  114. MedRouter: Demystifying Knowledge Differences Across Medical LLMs for Routing-Based ReasoningLLM reasoning → Clinical and biomedical applications
  115. Toward Comprehensive 3D Grounding: Orientation Grounding through Vision-Language ModelsMultimodal LLMs / vision-language models → Embodied and 3D vision-language models
  116. VPTwin: Real-Sim-Real Video Prediction for Robotic Manipulation PlanningWorld models → Embodied and robot world models
  117. Query, Align, and Distill: Navigation-Aware Cross-Modal Interaction for Efficient Vision-and-Language NavigationKnowledge distillation → Multimodal distillation
  118. OneSign: Unifying Sign Language Understanding Tasks with One ModelMixture-of-experts → Multimodal MoE models
  119. From Constitutions to Control: Interpretable Rewards for Aligning Language ModelsLLM alignment / RLHF / preference optimisation → Human values and AI feedback alignment
  120. SwingRL: Adaptive Observation Reinforcement Learning with World-Model Prediction for Cable-Suspended Hoisting ControlWorld models → Latent dynamics models and model-based RL +1
  121. SketchSSM: Write to the Full State, Read from a Compact SketchState-space sequence models (S4, Mamba) → Core SSM architectures for language +1
  122. Agent Safety From Within: Detecting Harmful Trajectories from LLM Internal StatesLLM agents and tool use → Agent threats, guardrails and governance
  123. Byzantine-Robust Federated RAG via Aligned Calibration and Fixed-Membership Conformal PredictionRetrieval-augmented generation → RAG robustness and adaptive retrieval
  124. Artificiality's First Exam: An LLM-Referenced C-test for Supraintelligence
  125. SemTree: Judge-Aware Semantic Decision Trees for Inspectable and Repairable Rule Decisions
  126. Enhanced Trading Systems with Intelligent Market Analysis
  127. GaussianBench: Physics-Fidelity Evaluation for Gaussian Scene Representations
  128. Is Word Error Rate Enough? Rethinking Privacy Evaluation in Speech with Entity-Aware Metrics
  129. Budgeted Cache Repair for Cross-Context KV-Cache Reuse
  130. TutlAit v1: a crowdsourced Moroccan Tamazight speech dataset with Arabic transcriptions and regional accent labels
  131. Who Warmed the Archives? LLMs Overestimate Historical Warmth
  132. The Rashomon Wikipedia: A Data-Perspectivist Analysis of Divergent Historical Narratives
  133. Larry Caused the Car to Stop, But the Model Didn't Notice: Transformer Blindness to the M-Heuristic
  134. Risk-Controlled Selective LLM Answering by Pricing Label-Free Checks
  135. Physics-Guided Flow-Map Matching for Precipitation Nowcasting
  136. PoE-Fuse: Precision-Weighted Expert Fusion for Bi-Temporal Change Understanding
  137. Probability Contracts: Accuracy, Coherence, and Decisions Across LLM Interfaces
  138. Event-Only Wingbeat Counting under Camera Motion: A Controlled MuJoCo Benchmark
  139. Engineering Efficient Self-Play Chess: Search, Replay, and Throughput Under Limited Compute
  140. Wasserstein Causal Forests for Distribution-Valued Outcomes
  141. Self-discovering RL in the Era of Experience: Is Learning History an Asset or a Burden?
  142. Learning in the Transverse Subspace: A Minimal Representation for Divergence-Free Operator Learning
  143. 3D Point Tracking with State Space Models
  144. ADPTNet: Adaptive with Prescriptive Timescales Non-Linear SSM for Sequence Modelling
  145. Re:Cognize -- Open-Set Comic Character Re-Identification
  146. Posterior Regimes and Latent Deception: Variational Bayesian Inference in Hidden Markov Models for Sequential Fraud Detection in Financial Transactions
  147. Uncovering shortcut learning in audio classifiers by discovering recurring concepts in temporal explanations
  148. Structure-Adaptive Tree Field Integrators
  149. Jev in Medicine: A Benchmark Evaluation
  150. SR4-Fit: A Unified Interpretable Rule-Based Machine Learning Framework for Informative and Trustworthy Decision-Making
  151. Estimate, Don't Imitate: Reusing Differentiable State-Based Policies for Visuomotor Control
  152. A Computer Vision Approach to Visual Fraud Detection in Phishing Websites Using YOLOv8
  153. LTV-CTDNet: Compositional Turning Decomposition for Short-Term Turning-Movement Forecasting
  154. When Known Physics Helps Neural PDE Models: Residual Constraints Out-Regularize Generic Priors for Nonlinear Dynamics
  155. ZeroBot: Learning from Scratch in Minutes with Generative Real2Sim
  156. EHRAdapt: Adapting Pretrained Language Models to Electronic Health Records with Semantic Priors for Rare Clinical Events
  157. T-SNN: Temporal Simplicial Neural Network for EEG Decoding
  158. UnfoldCRF: Structured Mask Refinement with Image-Conditioned Latent Regions
  159. A packet-level digital hardware twin for commissioning megahertz diagnostic edge AI and plasma control system integration in tokamaks
  160. ASTRA: ADMM-Accelerated Topology Reconfiguration for Dynamic Satellite Constellations
  161. A Multi-Dataset Benchmark of YOLO-Based Weed Detection in Precision Agriculture
  162. ICMAPE: In-Context Multiagent Pure Exploration
  163. The Privacy Fallacy of Crowdsourced Fine-Tuning: Extracting Proprietary Data via Topic-Based Poisoning
  164. From HL to H+L-1 Parameters: A Hankel-Toeplitz Forecaster for Long-Term Time Series Forecasting
  165. High-Level Text Preprocessing for Semantic Similarity Analysis of Discursive Texts: A Framework and Empirical Demonstration
  166. Future Information-Directed Sampling for Bayesian Nonstationary Bandits
  167. Adapting neural operators for mechanics decisions under changing operating conditions
  168. GroupMask: Layer-Adaptive Group-wise Sparsity for Semi-Structured LLM Pruning
  169. HARMONIA: Interpretable Graph Learning through Mixtures of Neural Bases
  170. Do System One Decisions Add Up? A Study of Probabilistic Coherence
  171. Gaussian Splatting-based Volumetric Video Compression with Sparse 4D Anchors
  172. Two-Sample Testing for Inhomogeneous Random Graphs in Non-Integral Lr Norms
  173. ThinkNet: Evolutionary Deep Ensembles for Efficient Motor-Imagery EEG Classification
  174. Greenpixie's AI Token Methodology: Assessing the Energy, Water and CO2-eq Impact of AI Tokens for Open and Closed Weight Models
  175. SymNetPro: LOS-Aware Directional Multi-Transmitter Localization from Sparse Radio Observations
  176. EpiTransfer: Sparse, Training-Free Long-Range Depth Estimation from Temporal Monocular Aerial Frames
  177. Test-Time Generalized Category Discovery
  178. How Strong Is the Evidence for the Artificial Hivemind? Reevaluating Evidence for the Open-Ended Homogeneity of Language Models
  179. Unifying Video Tasks via Spatiotemporal Analogy
  180. ArticulateArena: A Metric for Articulated Kinematics
  181. Diffusion-Based Rollouts as a Stabilization Mechanism for Long-Horizon Environmental Forecasting
  182. Preserving DEG Rankings for Gene Discovery in Histology-Based Spatial Gene Expression Prediction
  183. Optimizing the Phi-2 Small Language Model for Real-time Chatbot Applications Using Parameter-Efficient Fine-Tuning (PEFT) with QLoRA Quantization
  184. Quantization Error Is Spectrally Flat: A Single Random Probe Is a Calibrated, Data-Free Sensitivity Estimator, with Application to Budget-Targeted Mixed-Precision Quantization
  185. Validating Memory-Optimal Transformer Kernels on Real Hardware: From Formal Derivation to Measured Performance Across Two HPC Clusters
  186. Training Witnesses: Trusting the Training without Trusting the Trainer
  187. JIVE: Jacobian-Informed Volume Expansion for Diverse Generative Sampling
  188. SlopBench: How Well Can We Rank Language Models by Slop? A Multi-Domain Benchmark of Repetitive AI Writing
  189. On the Two Faces of Adam in Separable Linear Classification
  190. Finite Probes Suffice: Identifiability and Universality for Weight-Space Learning
  191. NSV-Shift: A Contrastive Benchmark for Non-Speech Vocalization Understanding and Response Adaptation in Speech-to-Speech Models
  192. No Free Efficiency: Revisiting the Trade-off Between Training Efficiency and Model Vulnerability
  193. MISHAP-Bench: A Hallucination Benchmark for Large Audio-Language Models
  194. Augmented Feature Boosting for Multicalibration
  195. Vanilla Policy Optimization Is Both Optimal and Differentially Private for Stochastic Contextual Bandits
  196. LLMs learn different forms of metacognition when trained to predict their own accuracy
  197. Lost with a Map: Conversational State and Behavioral Reliability in Language Models
  198. Neuron-Level Architecture Growth: A Controlled Evaluation for EEG Time-Series Decoding
  199. DCEmbed: Scalable Optimization over Neural Surrogates
  200. Curating Merchant-Matching Training Data with Two Confidence-Gated Local LLM Judges
  201. JET: Justification Evaluation in Transformer
  202. Robot-GST: geometry-aware spatial-temporal robot policy representation and evaluation
  203. Population Physics, Population Problems: Safety and Emergence in LLM Societies
  204. Conformal Coverage of Time Series: Validity and Inference
  205. Valid and Efficient Split Conformal Regression for Time Series
  206. In-Context Adaptation of Encoder-Decoder Models in Speech Recognition
  207. PI-NOMT: Physics-Informed Neural Optimal Mass Transport for Brain Fluid Dynamics
  208. Rethinking Contextualization by Reinterpreting Attention Head Channels
  209. How code helps different tasks? A decompositional lens on LLM post-training
  210. Laya as a Typed Probabilistic Assessor: An Independent Reproduction and a Preregistered Study of Calibration and Selective Escalation
  211. DeltaSeek: Toward Active Perception in Evolving Construction Environments
  212. An Active-Bottleneck Mechanism for Weak-to-Strong Generalization
  213. One Attack to Fool Them All: Highly Transferable Black-Box Adversarial Attacks on Frontier MLLMs
  214. dOPT: Differentiating Conic Optimization via Geometric Reduction
  215. Annealed Sinkhorn with Momentum: Certified Unregularized Optimal Transport in Linear Memory
  216. Eyes on the Road: A Naturalistic Comparison of MTW Rider Gaze in Urban Indian Traffic
  217. MinkowskiPE: Minkowski Positional Encoding for Spatiotemporal Perception
  218. Autonomous phase discovery
  219. Binding Multiple Modalities via Multimodal Wasserstein Barycenter
  220. RAISE: Reinforcing Access Control Policy Synthesis in LLMs via Symbolic Evaluation
  221. Multi-Aperture PPG with MAPIS: Spatial Optical and Temporal Coherence Fields
  222. LA-CPD: Local-Evidence-Aware Change-Point Detection for Human-LLM Authorship Segmentation
  223. Learning Strategies to Break Judges
  224. Skill2Env: Capability-Oriented Environment Synthesis from Skills for General Agents
  225. M3-Score: Fidelity, Memorization and Coverage as Separate Axes for Evaluating Generative Radiology Image Models
  226. Taming the Greeks: Option Portfolios with Inductive Biases
  227. Beyond Fixed Features: Architecture-Dependent Sensitivity to Node Representations under Heterophily
  228. StatD2GAN: When Calibration Masks Generator Quality in Held-Out Evaluation of Synthetic Weather Sequences
  229. Oracle-Efficient Online Classification with Stochastic Inputs and Adversarial Outputs
  230. ENet-GP: Unified Document Image Restoration
  231. YuE2: Unifying Symbolic and Audio Music Generation at Frontier Quality
  232. Transformer-based Neural Beamforming for Real-Time Speech Enhancement on Smart Low-Power Hearable Devices
  233. Collaborative Synthetic Data for Privacy-Preserving Financial Fraud Detection Across Organizational Silos
  234. ResDiffFRG: Residual Diffusion for Multiple Appropriate Facial Reaction Generation
  235. Task-Aware Discretization of Differentiable Logic Gate Networks
  236. PQ-HSA: Reusing Product-Quantized Scores for Hybrid Sparse-Approximate Attention
  237. Multi-Marginal Inverse Optimal Transport for Contrastive Learning Via Explicit Anchor-Positive-Negative Coupling
  238. Weighted Spline-Expanded Networks with Distributional Balancing for Continuous Treatment Effects
  239. The Effects of Incremental Instruction Delivery on Language-Model Creative Writing
  240. Yor`ub'a in Unicode: An Overview of a Problem
  241. HTN Planning as a Coordination Layer for Multi-Server MCP Tool Orchestration
  242. Robust Biomolecular Complex Design Across Protein Conformational Landscapes
  243. Reliable Replay through Spatial Coherence in Online Continual Learning
  244. BOReFT: Manifold Steering of Language Models for Black-box Optimization
  245. Shared Experience, Separate Learning: Companion Confidence Calibration for LLMs
  246. From Granular Revision Operations to Meaningful Revision Units: Evaluating LLMs for Revision Boundary Detection
  247. Revisiting Diffusion Fine-Tuning for Unsupervised Domain Adaptation
  248. Theory Guided and Interpretable Neural Operator Design for Partial Differential Equation Learning
  249. A Spectral Theory of Compositional Learning
  250. Non-Adaptive Learning of Sparse ErdHos--R'enyi Graphs via Affine Splitting
  251. Understanding Confabulation and Rethinking Reconstruction in Activation Explanations
  252. SpecRead: A Benchmark for Measuring Whether Language Models Understand Hardware Specifications
  253. Geometric Inductive Biases for Semi-Supervised Equalization: The Constellation-Aware Transformer
  254. Dynamic Kuramoto-Hodge Operators for PDEs on Complex Geometries and Topologies
  255. TopoMamba: A Load-Support Relation-Guided Multi-Directional State-Space Model for Topology Optimization
  256. Resource-Aware Parameter-Efficient Model Adaptation for Onboard High-Dimensional Data
  257. T-MoXAI: A Hierarchical Explainability Framework for Temporal Multimodal Data
  258. MAD-Guard: Controlled Study of Autoregressive Generation versus Direct Decision Interfaces for Closed Multimodal Forensic Tasks
  259. Benign Overfitting for General Norms and Distributions
  260. Reachability is not enough: Diagnosing long-range behavior in GNNs
  261. Reset Is Not Recovery: Evaluating Recoverability from False Conversational Context via Sycophancy Hysteresis
  262. Closing the Cross-Dialect Gap: Query Plans as a Portable Interface in Text-to-SQL
  263. RSD-Poker: Structure-Adaptive and Shift-Robust Risk-Utility Certification for Residual Policies in Imperfect-Information Games
  264. When Noise Meets Long-Tail: Feature-Threshold Dual Calibration for Robust Pseudo-Labeling
  265. Scalable Attribution and Control of Model Behavior During Training
  266. One Model Is Not a Crowd: Multi-LLM and Aspect-Conditioned Diverse Comment Generation
  267. CompoWorld: Compositional Environment Scaling for General Agents
  268. LieDiscover: Adaptive Symbolic Library Construction for Explicit Open-form Symmetry Discovery
  269. Audit-First VAPO: Risk-Certified Selective Updates under Imperfect Verification
  270. FuseAlign: Forced Alignment in the Wild
  271. From Distributions to Stochastic Processes: Neural Approximation of Measure-Valued Maps
  272. Probe to Act: Elevating Browser-Use Agent via Active Visual Probing
  273. Learning to Learn from Context: Synthetic Training from Perturbed Public Documents
  274. Scalable and Data-Driven Decision Support in the Maintenance, Repair, and Overhaul Process
  275. Quantifying Behavioral Tails in Black-Box Language Models
  276. lapanda: A Matrix-Free Differentiable Solver for Nonconvex Constrained Optimization Layers
  277. Learning Dynamics of Continual Learning: A Unified View of Data Attribution, Forgetting, and Plasticity Loss
  278. Anguinus Sculpturae: Compositional Synthesis of Peak-Enhancement Breast DCE-MRI Scans
  279. Supervision Recovery for Time Series Anomaly Detection via Context-Anchored Pairing
  280. You Only Edit Once: Incentivizing In-Context Capability of LLMs via Local Demonstration Refinement
  281. Learning Transferable Reaction Mechanisms from Visual Chemical Knowledge
  282. Hierarchical Response Preservation for Continual Adaptation of Zero-Shot Graph-Text Models
  283. Short-Length Code Designs for Integrated Sensing and Communications: A Deep Learning Approach
  284. ViCoR: Reliable Molecular Structure Extraction via Spatially Aligned Verification and Executable Revision
  285. JustQuant: You Don't Need Smoothing, SVD, or Rotation for 4-Bit Activation Quantization
  286. The Price of Peeking: Anytime-Valid Leakage Detection on ML-KEM EM Traces
  287. LoopLUT: 3D Lookup Tables with Progressive Region Refinement for Real-Time 4K Image Enhancement
  288. Pretraining Transformers with Quantized Softmax in Attention
  289. TGRL: Temperature-Grouped Reinforcement Learning for Efficient Exploration in LLMs
  290. Approximating Softmax in Pretrained LLMs: Model Sensitivity and Kernel Acceleration
  291. Compressing Value Predictions for Learning-Augmented Metrical Task Systems
  292. OpenFC: Learning Verification Policies towards Open-Search Fact Checking
  293. HiLoRe: What to Store, Compress, or Recompute for Efficient GRPO Training
  294. Terminal-Register Certification for Finite-Measurement Learning of Multiscale Quantum States
  295. Correct then Forecast: Observer State-Space Models for Time Series Forecasting
  296. Dr. Free: You Don't Need Difficulty Rewards for Self-Evolving Search Agents
  297. GraphSelect for Budgeted Representation Selection in Multimodal Graph Inference
  298. Fine Until Fine-Tuned: Repeated Solutions Make Reasoning Fragile
  299. PGL-3D: Towards Progressive Geometric Learning for 3D Visual Query Localization
  300. Sparsity by Default: The Theory and Practice of ARD in Gaussian Process Regression for Variable Selection
  301. Calibrated Derivative-Process Sensitivity for Gaussian-Process Variable Selection
  302. TerMeZO: Ternary Sparse Zeroth-Order Optimization for Fine-tuning BitNet Models at the Edge
  303. OSCC: Certified Observation-Safe Coupling Optimization for Gradient-Noise Control in Imperfect-Information Learning
  304. Jev Matches 7B Language Models for Speech-Neuroprosthesis Rescoring
  305. ManiEdit: Sequential Unstructured Knowledge Editing for Language Models from a Manifold Perspective
  306. Neural Scaling Laws of Transformer Operator Network
  307. E-CONAN (Entailment, CONtradition And Neutral) Diagnostics Dataset Investigating Linguistic Phenomena in Arabic Natural Language Understanding
  308. Discovering Symmetries in Neural Network Parameter Spaces
  309. EverMine: Dissecting the Self-Evolution of Research Capabilities in Long-Horizon Alpha Research
  310. AI-Driven Collaborative Assembly Line Inspection: System Integration and Deployment Challenges
  311. LLM4Trust: Exploring the Capabilities of Large Language Models for Trust Evaluation
  312. Anatomy-Structured Hierarchical MIL for Weakly-Supervised Thoracic Disease Detection in Chest X-rays
  313. PPG-LM: A Photoplethysmography-Language Model with Multi-Level Clinical Alignment
  314. SLP-ProbHard: Probabilistic Hard-Constrained Learning via Structural Latent Parameterization
  315. Printability-Constrained Adversarial Decals for Near-Nadir Aerial Perception: Measured Ink Gamuts, Nested Realism Constraints, and a Physical-World Bound
  316. What Happens During Autonomous Deep Research After the User Steps Away?
  317. When Evidence Changes the Subject: Subject-Typed Claim Licensing for Learned Routing
  318. Pulseflow: PPG Counterfactual Generation Via Latent Transport
  319. The cost of useful natural gradient updates
  320. FedMHAR: Federated Multimodal Human Activity Recognition using Multi-Agent Reinforcement Learning
  321. Predicting Block-Coordinate Performance via Cross-Curvature
  322. What masking geometry works best for EEG foundation models?
  323. Mask-Induced Displacement in Audio XAI via Logit Trajectory Decomposition
  324. How Synthetic Labels Improve Conformal Prediction: A Perspective on Conditional Coverage
  325. Just Let Linear States Forget the Distant Past: Prefix Caching via Suffix Replay for Hybrid LLMs
  326. Geometric Identification in Predict-Then-Optimize Learning
  327. A Cheap Verifier is Good Enough: LLM Post-training is Robust to Erroneous Rewards
  328. GSM: Efficient Language Modeling with Shared Global State
  329. Rethinking Token Reweighting for SFT: Suppress, Reverse, and Extrapolate Learned Features
  330. Protected Cores Are Not Enough: Certifying AI-Proposed Revisions of Temporal Specifications
  331. When Does the Concept of "Dog" Emerge in an Audio LLM?
  332. A Free Knob: Decoupling Calibration and Predictive Skill in Threshold-Based Evaluation
  333. Sharp training-conditional coverage for conformal prediction under covariate shift
  334. Native Association: Confidence-Aware Human Perception in the Wild with a Foundation VLM
  335. A Visual Classification Dataset and Model Evaluation for Historical Manuscript Illustrations
  336. From PDF to Evidence: Structure-Aware Retrieval for Clinical Practice Guidelines
  337. HESP: Separating What to Probe from When to Stop in Local LLM Alert-Triage Agents
  338. Context Spanning: A Communication Framework for Full-Duplex Speech Models and External LLM Backends
  339. MIC: Explaining Image-Claim Inconsistencies in AI-Generated Multimodal Misinformation
  340. MAC-Net: A Multi-Task Deep Learning Framework for Modeling Cognitive Function From Task-Based fMRI
  341. SMAT: Simple and Efficient Merge-Aware Training
  342. SchemaMem: Schema-Indexed Recurrent Memory for Delayed State Retrieval
  343. CORA: A Protocol for Diagnosing Boundary Robustness in Text-to-Audio Retrieval under Query Reformulations
  344. TC-ADA: One-Shot Active Domain Adaptation for Semantic Segmentation
  345. Temporal Graph Learning of Wearable Actigraphy and Sleep Traces for Modelling Adolescent Crystallized Intelligence
  346. A Light Bilevel Refinement Aligns Self-Supervised Representations for Stronger Task-Specific Learning
  347. Explainable Deep Learning of Resting-State Functional Connectomes Reveals Network Biomarkers of Adolescent Intelligence
  348. Recovering Lower-Dimensional Semialgebraic Support of a Measure from its Moments
  349. TT-VidT: Decoupling the Temporal Axis for Efficient Motion-Centric Video Pretraining
  350. Grounding Memory Summarization in Utility Intent
  351. Investigating the Effect of k-NN Preprocessing on Developing Graph Neural Networks: A Fairness-Based Perspective
  352. MetaBench-Harness: Unlocking End-to-End Optimization of Benchmark Harnesses
  353. Let CSP Be Your ANCHOR: Adaptive Crystal Search over Frozen Structure Priors
  354. Decoupling Token Roles in Autoregressive Pretraining
  355. VaME: Exploring Variational Latent Reasoning for Multimodal Embeddings
  356. SciGen-Verifier: A Multimodal Reasoner for Explainable Verification in Scientific Image Generation
  357. Preserving Morphemes: Morphology-Guided Pre-Tokenization for Nepali
  358. Cross-modal Translation via Conditional Latent Denoising for Video Deepfake Detection
  359. AutoHGNN: Robust and Efficient Neural Architecture Search for Hypergraph Neural Networks
  360. From Grey-Box to Green-Box: When can Physics-Informed Machine Learning Reduce Carbon Footprints in Structural Health Monitoring?
  361. Local LMO is Secretly a Projection Method!
  362. Recursive Harness Distillation across Agents for Robot Manipulation
  363. Optimal Transport Dropout for Structured Predictive Uncertainty
  364. TNF based Spectral Embedding for Effective Application of Supervised Machine Learning Techniques in Automobile Insurance Fraud Detection
  365. What Survives the Codec Shift: Pooled No-Vocals Residuals for Speech Deepfake Detection
  366. Identifying the Predictable Drift of a Semimartingale from Marginal Laws
  367. The Selection Rule Decides the Winner: A Pre-Registered Audit of Open-Set Graph Anomaly Detection
  368. DrafTS: Time-Aware Decomposition with Residual Correction for Time Series Modeling
  369. Geometry-Adaptive Mechanisms for Private Synthetic Data
  370. From Script to Drama: An Agentic Framework for Controllable Multi-Speaker Dialogue TTS
  371. When Does Geometric View Synthesis Help Wine Label Retrieval? A Public One-Shot Benchmark Across Self-Supervised and Vision-Language Backbones
  372. DISCERN: Can AI Agents Work Like Scientists and Guide Discovery?
  373. Long-Horizon Analog Design Bench: Benchmarking Agents on Hours-Long Analog and Mixed-Signal Circuit Design Tasks
  374. Focus and Supplement: Dual-Enhanced Vision Transformer for Multi-Class Anomaly Classification
  375. How Much Imprecision is Enough Imprecision in my Classifier? A Practical Elicitation Procedure
  376. QuPID: Quantum Parameter-Efficient Input-Dependent Retrieval Adaptation for Medical RAG
  377. KoopCell: Koopman-Based Generative Model for Learning Single-Cell Dynamics from Distribution Snapshots
  378. Language Discrimination Improves Linguistic Learning in Multilingual Speech Models
  379. ReLoc: Rethinking Scene Coordinate Regression Architecture for Robust Outdoor LiDAR-based Localization
  380. CHI: A Composite Hallucination Index Unifying Entity, Relation, and Quantity Dimensions for Summarization Evaluation
  381. CalibHyper: Chance-Corrected Relational Hypergraphs for Few-Shot Molecular Property Prediction
  382. RAEGL: Risk-Aware Evidence-Gated Learning for Selective Contextual Routing under Temporal Shift
  383. Naturalness-guided Manifold Flow Matching for Sign Language Production
  384. FeCoSplat: Feedback-Guided Compression for Feed-Forward 3D Gaussian Splatting
  385. ANTMAN: Adaptive Need Tracking for Multi-Agent Navigation in Large Information Spaces
  386. VisionHOPE: Visual Backbones as Self-Modifying Learning Systems
  387. PhysAlign: A Benchmark for Evidence-Grounded Role Alignment in Multimodal Physics Reasoning
  388. PIC-UIE: Predicting Image-Adaptive Corrections for Lightweight Underwater Image Enhancement
  389. ZeroGAR: Benchmarking the Adversarial Robustness of Zero-Shot Graph Models
  390. When Privacy Moves ML-Mediated Decisions On Device: Information and Incentive Misalignment in Auctions
  391. SocialHumanoid: Towards Expressive Humanoid Behavior via One-Step Co-Speech Motion Generation
  392. LoopTrack: A Simple Baseline for Parameter-Efficient Transformer Tracking
  393. BITS: Rethinking Fair and Comprehensive Evaluation for Irregular Time Series Forecasting
  394. BaatCheet: A Multilingual Corpus for Dialogue Translation in Indian Languages
  395. Informative Viewpoint Selection for Episodic-Memory Embodied Question Answering using Omnidirectional Images
  396. Feedback Makes Perfect: A Closed-Loop Framework for NL-to-STL Translation
  397. InfoEdit: Probing Global Layout Reasoning in Infographic Editing
  398. RINI: Seeing the Prior Is Not Enough
  399. Multi-Dimensional Comparative Scale Construction for Efficient Personalized Subjective Judgment in High-Traffic Applications
  400. The Text Beside the Image: Detection, Utility and Leakage for Trustworthy Multimodal Medical Data and Beyond
  401. Domain Generalization under Sampling Pattern Shifts in Irregular Time Series
  402. Modelling non-linear aeroelastic loads in long-span bridges with extreme learning machines
  403. Towards Identifiable Representations under Misspecified Structure
  404. Next Thoughts Are Distributions: Generative Autoregressive Reasoning in the Latent Space
  405. Structured Sparse Memory for Recurrent Reasoning
  406. LSTMem: Hierarchical Long Short-Term Online Memory for Large Language Models
  407. SCISSOR: Score-Conditioned Instrument Source Separation for Orchestral Recordings
  408. EngIntervene: Benchmarking Multimodal Engineering State Understanding and Design Intervention Reasoning
  409. CORTEX: A Verified Experience Layer for Generalist Agents
  410. PORTER: Edge-Cloud Residency for Persistent 3D Scene Graph Memory
  411. Two Heads Are Better Than One: Aggregating Weaker LLMs for Better Forecasts
  412. BERT4DTI : BERT-based Model for Predicting Drug-Protein Interactions
  413. Feedback-Robust AI for Patient Knowledge Graphs
  414. The Price of Locality: Why Forward-Forward Underperforms Backpropagation?
  415. Inspire: Benchmarking Scientific Literature Search for Open Research Problems
  416. MTLiquid: Enabling Efficient Multi-Task Learning using Liquid Neural Networks for Lightweight Healthcare Monitoring Systems
  417. AevaScenes: An FMCW LiDAR Dataset and Benchmark for Long-Range Perception
  418. RepFlow: Reciprocal Supervision Improves Generation and Representation in Flow Models
  419. CoLMbo-SV: A Grounded Language Model for Explainable Speaker Verification
  420. Beyond Calibration: Do a Typed-Decision Model's Probabilities Obey the Probability Axioms?
  421. MorphAtt: A Neuromorphic Accelerator for Efficient Multi-Head Attention Processing in Spiking Vision Transformers
  422. SMORE: Stability-Promoting Mesh-Agnostic Model Reduction for Time-Dependent PDEs
  423. Turning Speech Language Models into Multilingual Listeners
  424. Are Benchmarks Reliable? Toward Structural Diagnosis via Item-Level Capability Boundaries
  425. Orthogonal Witness Control for Muon Optimization via Sigmoid Spectral Reshaping
  426. Apparent Compression, Real Stability: The Intrinsic Dimension of Learning a Quantum Wavefunction
  427. AG-CoT: Verified Algorithmic Traces for LLM Program Synthesis on Clifford Circuits
  428. When an Evaluation Rule Writes Training Labels: Measuring Human-Reference Forgiveness in NAVSIM
  429. Unlocking Latent Personalization in LLMs
  430. Which Self-Improvements Should We Trust? Reliable Self-Improvement When Agents Reuse Their Benchmarks
  431. Scoring the Wrong Question: Readout Failures in Constrained-Option Evaluation
  432. Can Protein-Derived Knowledge Improve Pathology Foundation Models?
  433. ABO-Med: Accelerated Bilevel Optimization for Few-Shot Medical Image Classification
  434. When Does Backpropagating Through Policy Memory Matter? Physical Credit, Optimizer Updates, and Observability
  435. FloodDiffusion 2: Efficient and Path Controllable Streaming Motion Generation
  436. Beyond Tasks: A Vision for Reproducing an Animal-like Behavioral Substrate Using Modern Robot Learning Techniques
  437. Sharp Critical Minimax Laws and No-Learning Thresholds in Continuous-Time Adaptive Control
  438. Convergence of Practical Muon
  439. Generalization Dynamics of LM Pre-training
  440. CFLoRA: Federated Fine-tuning of LLMs with Complementary Factors for Error-free Aggregation
  441. CAME: Company-Aware Evidence-Memory Experts for Interpretable Quarter-Ahead Revenue Forecasting
  442. Knowing Is Not Choosing: What Explicit Verification Adds Beyond Generative Preference
  443. Ceiling of a Task: When Can a Transformer Succeed Without Its Chain of Thought?
  444. Quantum Monte Carlo Tree Search with Fixed Confidence
  445. ILP-BO: Integer Linear Programming-Based Black-Box Optimization
  446. Save Your Saturated Data: Learning Beyond Reward Saturation in Group-Based RL
  447. Compositional Safety Failures in Harness Evolution: Identification and Runtime Monitoring
  448. Classifying Dominant Temporal Orientation without Pretrained Text Embeddings: A Novel Morphosyntactic Inventory Vector Approach
  449. Mycelium: A Generalizable Cross-Grid Multi-Task Model for Electrical Distribution Systems
  450. ECG-Scroll: A Long-Horizon, Streaming Benchmark and Agent Environment for Interpretation of Ambulatory Electrocardiograms
  451. Modular Discovery of General Game-Playing Algorithms with Large Language Models
  452. Can Tabular Foundation Models Amortize Statistical Inference?
  453. ParallelPilot: Supporting Coordination and Monitoring in Parallel AI Coding
  454. Simulation-Free Learning of GP-SDEs from Irregular Observations
  455. D-JEPA: Design-Recoverable JEPA Representation with Swappable Physics Decoders
  456. CARVE: Breaking Data Barriers in Chip Placement by Harnessing Reusable Expertise
  457. Evolving Dexterous Robots from Scratch
  458. Octree-based Video Representation
  459. How Linear Attention Remembers
  460. Geometry-Aware Operator Families for Structured Representation Learning
  461. QSCP: Beyond Class-Name Prompts for Query-Guided Semantic Change Parsing
  462. Plug-and-Play Methods Provably Converge Even with Improperly Trained Denoisers: Convergence by Architectural Design
  463. The Model Knows Another Way: Strategy Switching for Effective RLVR Exploration
  464. dKFD: Phase-Structured Evidence Allocation for Fixed-Budget Localized Event Understanding
  465. SemReward-VL: Semantic Reward-Guided Video-Language Adaptation for Developmental Behavior Assessment
  466. Structure-Mapping-Guided Self-Explanation for Learning Mathematical Procedures
  467. The limits of exactness: On the failure of automatic differentiation in physics-informed machine learning
  468. Parameter-Efficient 3D Segmentation of Liver and Liver tumors: Depthwise factorization Scales Better Than Dense Convolution with Spatial Dimensionality
  469. NutriVision: Ingredient-Conditioned Fusion and Prediction for Single-Image Food Nutrition Estimation
  470. QureRadEmbed: Structuring Radiological Similarity through Attribute and Reasoning Supervision
  471. KernelZero: Co-Evolving Proposer and Coder for Continuously Improved GPU Kernel Generation
  472. Algorithmic Harms Associated with Generative Model-Augmented Recommendation Systems
  473. Zero-Storage Procedural Neural Synthesis via Boundary Dynamics: Formal Verification in Lean 4 and Bare-Metal Gauntlet Validation
  474. Reading Too Much into Context: Passive Exposure Can Steer LLM Decisions
  475. LLM sequential decision making under uncertainty in biochemical domains
  476. Deep Learning Techniques for Phoneme Recognition in Italian Children' s Speech
  477. Large Language Models Substantially Compress Well-Being Inequality but Largely Preserve Its Socioeconomic Structure
  478. BudgetVerify: Budget-Tiered Verification for Financial QA
  479. DevelopmentODE: Structured Neural ODEs for Early Brain Development Dynamics Across a Decade
  480. Cost-free Spectral Estimation for Adaptive Newton--Schulz in Matrix Optimizers
  481. Pinned and Still Unstable: Within-Judge Verdict Variance and the Noise Floor of LLM-as-Judge Leaderboards
  482. More than 83.69% of the zeros of the Riemann zeta function are distinct
  483. Balancing Early Performance Sacrifices with Long-Term Gains: Scaling Learning-Rate Warmup Duration Across Training Horizons
  484. Improving the Diversity of LLM Outputs without a Trade-off