mlsanityData to 14 Oct 2026

Papers

New on arXiv

Wed 30 Sep 2026

937 papers

  1. Jev-as-a-Judge for RAG Claim Verification: Confidence-Gated Cascading Between a Non-Generative Verifier and a Frontier LLMRetrieval-augmented generation → RAG robustness and adaptive retrieval
  2. MeddeX: A Retrieval-Augmented LLaMA-3 System with Gemini Vision for Evidence-Backed Medical Report AnalysisRetrieval-augmented generation → Medical and biomedical RAG
  3. Beyond the Sycophancy Score: How Task, Model, and Pressure Shape LLM YieldingLLM safety and security → Safety alignment, refusal and guardrails
  4. Component and Dimension Sparsity in Transformer Refusal MechanismsMechanistic interpretability → Activation steering methods
  5. Tree Navigation Without LLM Summaries: A Matched-Cost Study of Hierarchical Retrieval for Long-Document QARetrieval-augmented generation → Knowledge-graph and structured-knowledge RAG
  6. What Do Verifiable Rewards Teach Video-Language Models About Time? A Controlled Multi-Model StudyLLM reasoning → Reinforcement learning for reasoning
  7. OceanMind: A multi-agent AI system for ocean diagnosisLLM agents and tool use → Web, search, and research agents
  8. MintFlow: Minimal Trajectory Intervention for Constrained Flow MatchingDiffusion models → Flow matching, time series and science
  9. Rank-Aware Speculative Sampling for Diffusion Draft TreesDiffusion models → Diffusion foundations and sampling +1
  10. State-Space Unlearning for Non-Stationary Bias in Land Surface ForecastingState-space sequence models (S4, Mamba) → Time series forecasting with SSMs
  11. Toward Controlling Biology with Language:Offline Learning of Prompt-Conditioned Interventions for Cells, Organoids, and BiobotsMultimodal LLMs / vision-language models → Embodied and 3D vision-language models
  12. SHARPO: Segment-Level Credit Assignment for Agentic Reinforcement LearningLLM agents and tool use → Reinforcement learning for LLM agents +1
  13. Kepler: Auditable World Models for ARC-AGI-3World models → World models for language agents
  14. Align Then Reason: A Multimodal Lip-Sync Judge for DubbingLLM reasoning → Audio-language model reasoning
  15. Training-Aware Target Coverage for Synthetic Data SelectionSynthetic data and self-improvement for LLMs → LLM synthetic data generation
  16. Paying for Too Many Tokens? Valid and Cost-Efficient Multimodal LLM Annotation with Simple HeuristicsMultimodal LLMs / vision-language models → Video understanding with MLLMs
  17. VTV-FM: Flow Matching through Variational Terminal-Velocity ClosureDiffusion models → Flow matching, time series and science
  18. Effective Synthetic Data Curation Requires Group-Level SignalsSynthetic data and self-improvement for LLMs → Model collapse from synthetic data
  19. Learning Goal-Reaching Quasimetric Geometry From Finite-Time ReachabilityDeep reinforcement learning methods → RL theory and sample complexity
  20. Scalable Multi-Task Inverse Reinforcement LearningOffline RL and imitation learning from data → Inverse reinforcement learning +1
  21. Video Evidence Indexing: Learning Where to Look from Video Previews for Token-Budgeted Long-Video Question AnsweringKnowledge distillation → Multimodal distillation
  22. Benchmarking Generative Models for Weather Data Assimilation on Real Station ObservationsDiffusion models → Flow matching, time series and science
  23. CF-JEPA: Improving Robustness of JEPA World Models via Controllability FactorizationWorld models → Latent predictive world models
  24. Reason in Style: Discovering and Controlling Style in Language ModelsLLM reasoning → Mathematical reasoning
  25. JEPA-TTT: Persistent Test-Time Training of Latent World Models for Planning under Dynamics ShiftsWorld models → Latent predictive world models
  26. Robust Nash Alignment under Preference UncertaintyLLM alignment / RLHF / preference optimisation → Reward modeling and RLHF
  27. ReLiveGym: Evaluating Long-Lived Agents over Weeks of Replayed RealityLLM agents and tool use → General LLM agent frameworks and benchmarks
  28. Meta-Multi-Agent Reinforcement Learning for Fast Adaptation of Interactive Policies with Applications to Autonomous DrivingEnd-to-end autonomous driving → Reinforcement learning for driving
  29. SkillSpec: Consensus-Gated Agent Skill Evolution via Representation SpecializationLLM agents and tool use → Agent skill learning and evolution
  30. SemanTok: Predictable Semantic Tokens for Efficient Autoregressive Video GenerationWorld models → Video-generation world models and simulators +1
  31. Backdoor Purification for LoRA-Tuned LLMs via Null-Space ProjectionLLM safety and security → Backdoor and data poisoning attacks
  32. Ontology-Grounded, Reasoner-Verified Benchmarks for Evaluating LLM Reasoning in Scientific AILLM reasoning → Knowledge- and tool-augmented reasoning
  33. VisionQ: VLM-as-a-Judge Taxonomy, Dataset and Benchmark for Qualitative Analysis in Computer VisionMultimodal LLMs / vision-language models → VLM benchmarks and visual reasoning
  34. Backdoor Containment via Expert Quarantine and Shutdown in LLMsLLM safety and security → Backdoor and data poisoning attacks
  35. Exploring More, Reasoning Better: Stepwise Risk-Sensitive GRPO for Diffusion Language ModelsLLM reasoning → Diffusion reasoning and test-time scaling +1
  36. Agent Evaluation Reliability: More Tasks Won't (Always) Fix An Agent LeaderboardLLM agents and tool use → General LLM agent frameworks and benchmarks
  37. Self-Evolving Coding Rules for AI Coding AgentsLLM agents and tool use → Software engineering coding agents +1
  38. Incident-Arena: Getting agents to the last nine of reliabilityLLM agents and tool use → Software engineering coding agents
  39. TacDyn-WAM: Learning Implicit Tactile Dynamics in a Heterogeneous Visuo-Tactile World Action ModelWorld models → Embodied and robot world models
  40. CompMat-Bench: Benchmarking AI Agents for Computational Materials ScienceLLM agents and tool use → Web, search, and research agents
  41. HAWK: Rethinking Multimodal Drafting for Speculative DecodingLLM reasoning → Multimodal reasoning +2
  42. Beyond Supra-Competitive Outcomes: Collusive Behaviour in Deep Reinforcement Learning for Optimal Execution GamesDeep reinforcement learning methods → Adversarial robustness and RL in finance
  43. Spatial Strategies, Not Actions: Vector-Quantized Geodesics as Tools for LLM-Driven AgentsLLM agents and tool use → Multimodal and embodied agents
  44. Legal Research Bench: Measuring End-to-End Reliability in Long-Horizon Legal Research AgentsLLM agents and tool use → Web, search, and research agents
  45. MIKASA-Robo-VLA: Benchmarking Memory in VLA Models for Long-Horizon ManipulationVision-language-action models → Failure recovery and robustness
  46. When Reasoning Helps Action: Monitoring and Steering Chain-of-Thought in Vision-Language-Action PoliciesLLM reasoning → Prompting and chain-of-thought reasoning +1
  47. ALER: Adaptive Learnable Experience Rewriting for Reinforcement LearningDeep reinforcement learning methods → Deep RL algorithms and representations
  48. Discrete Annotation, Continuous Preference: Rethinking Supervision for Accurate and Generalizable Aesthetic Image CroppingLLM alignment / RLHF / preference optimisation → Diffusion and multimodal model alignment
  49. Gestalt: Large Multimodal Interplay ModelMultimodal LLMs / vision-language models → MLLM training and visual instruction tuning
  50. Token-World: World Modeling in Vision-Language Model Token Space for Robot ManipulationWorld models → Embodied and robot world models +2
  51. FORTE: Adaptive Scoring and Exact Keyframe Selection for Long-Video Question AnsweringMultimodal LLMs / vision-language models → Video understanding with MLLMs
  52. Interpreting Reasoning of Large Language Models via Partial Information DecompositionLLM reasoning → Test-time compute and efficient thinking
  53. Emergent Unfaithfulness: How Alignment Training Causes Language Models to Silently Override Task FaithfulnessLLM alignment / RLHF / preference optimisation → Human values and AI feedback alignment
  54. Can LLMs Reason Over Long Horizons? An Empirical Evaluation of Context Strategies for Longitudinal Clinical ReasoningLLM reasoning → Clinical and biomedical applications
  55. PhysVista: Benchmarking Physical Intelligence in VLMs via a Perception-Reasoning-Assessment LoopLLM reasoning → Spatial and embodied reasoning +1
  56. Redundancy Meets Synergy: Dependency-aware Expert Selection for MoE via Submodular OptimizationMixture-of-experts → MoE theory and gating functions
  57. No One Architecture Fits All: A Cross-Environment Evaluation of Hierarchical Red Team AgentsLLM agents and tool use → Reinforcement learning for LLM agents
  58. Does Continual Imitation Learning Remain Grounded? A Language-Perturbed Benchmark for Robotic Task RetentionOffline RL and imitation learning from data → Language-conditioned policies and VLAs
  59. Same Scene, Different Task: Skill Alignment for Compositional Generalization in VLAsVision-language-action models → Generalist VLA models and benchmarks
  60. Before Agents Decide: Epistemic Action in LLM-Based SystemsLLM agents and tool use → Social simulation and multi-agent debate
  61. Denoising Surface: Modeling and Predicting Inference Cost for Diffusion LLM ServingEfficient LLM inference and serving → Diffusion LLM and MoE inference
  62. Score the Update, Not the Token: Descent-Aligned Routing for Combinatorial LoRA ExpertsMixture-of-experts → Expert routing strategies
  63. EurekaBench: Measuring Agentic Ability to Discover New Scientific InsightsLLM agents and tool use → Web, search, and research agents
  64. PixelDense: Dense Prediction as Representation Alignment for Pixel DiffusionDiffusion models → 3D, point cloud and scene generation
  65. JevSpawn: Adaptive Agentic Inference through Compositional Action SpacesLLM reasoning → Reasoning for agents, planning and code +1
  66. How AI Agents Discover Scientific Equations: From Hydrotope Rediscovery to New Water-Wave AmplitudesLLM agents and tool use → Web, search, and research agents
  67. Memetic Trojans: Social Contagions as Carriers of Adversarial Payloads in Agent NetworksLLM agents and tool use → Agent threats, guardrails and governance
  68. Code That Works, Environments That Don't: Measuring Environment Reproducibility in AI-Generated SoftwareLLM agents and tool use → Software engineering coding agents
  69. CommunityKV: Efficient Long-Context Decoding via Graph PartitioningEfficient LLM inference and serving → KV cache optimization
  70. Source Identification Is Not Fitness Testing: Measuring the Limits of Synthetic-Data AttributionSynthetic data and self-improvement for LLMs → Model collapse from synthetic data
  71. EchoPress: Query-Agnostic KV Cache Pruning via Virtual Context ReconstructionEfficient LLM inference and serving → KV cache optimization
  72. LLM-as-a-Judge for Low-Resource Languages: Adapting Ragas and Comparative Ranking for RomanianRetrieval-augmented generation → RAG pipeline design and evaluation
  73. Representation Transitions Reveal Emerging Safety Risks in Multi-Turn LLM AgentsLLM agents and tool use → Agent threats, guardrails and governance
  74. NEUROTOKEN: Joint Source and Directional AAD with Envelope Decoding via Conditional Flow MatchingDiffusion models → Audio, speech and music
  75. Specificity-Aware Diffusion Steering via Variance-Reduced Sequential Monte CarloDiffusion models → Diffusion foundations and sampling
  76. T2SPO: Trajectory-to-Step Policy Optimization for Agentic Reinforcement LearningLLM agents and tool use → Reinforcement learning for LLM agents
  77. When Harnesses Lose the Signal: Causal Evaluation of Recovery in LLM AgentsLLM agents and tool use → Reinforcement learning for LLM agents
  78. DeepJEPA: Scaling World Models from WithinWorld models → Latent predictive world models
  79. MoRA: MoE Pruning via Router Bias Learning and Expert ApproximationMixture-of-experts → Expert routing strategies
  80. Manifold-Constrained Initial Noise Optimization for Efficient Generative Model AlignmentLLM alignment / RLHF / preference optimisation → Diffusion and multimodal model alignment
  81. Diffusion Editing with Soft Mask: Pixel Level Redo of Image and Video with Adjustable StrengthDiffusion models → Image generation, text-to-image and editing
  82. Ranking-Aware Prompt Optimization for Multimodal Clinical DiagnosisMultimodal LLMs / vision-language models → Medical vision-language models
  83. Semifactual Credit-Augmented Policy OptimizationLLM reasoning → Reinforcement learning for reasoning
  84. Physis-Lang: Self-Evolving Language as a Physical Representation for Video World ModelWorld models → Video-generation world models and simulators
  85. WorldAuditBench: Interactive 3D World Auditing with Multimodal AgentsLLM reasoning → Spatial and embodied reasoning +1
  86. Cogentic: Multi-Agent Orchestration for Automated Proof DiscoveryLLM agents and tool use → Educational agents and math problem solving
  87. Scaling Laws for Looped Mixture of ExpertsMixture-of-experts → MoE theory and gating functions
  88. Looped Diffusion TransformerDiffusion models → Diffusion foundations and sampling
  89. How Much of a Harness Does a Strong Agent Need for Autonomous ML Engineering?LLM agents and tool use → Software engineering coding agents
  90. Linguistic Loopholes in LLM Unlearning: From a 174-Language Benchmark to Coverage-Aware UnlearningLLM safety and security → Machine unlearning
  91. PivotOPD: Learning to Recover from Pivotal Mistakes in Multi-Turn AgentsKnowledge distillation → On-policy and self-distillation for LLMs
  92. cua-speedrun: Standardized Benchmarking of the Speed of Computer-Use AgentsLLM agents and tool use → GUI and mobile agents
  93. OpenTSLM TeeMoE: A Unified Time-Series Language Model for Forecasting, Contextual Prediction, and ReasoningLLM reasoning → Time series reasoning
  94. ComputerSD: Online Self-Distillation from Real-Time Feedback for Computer-Use AgentsLLM agents and tool use → Reinforcement learning for LLM agents +1
  95. Distribution Matching Distillation for Continuous Diffusion Language ModelsDiffusion models → Discrete and text diffusion
  96. EviRover: Reinforcing Agentic Perception Beyond a GlanceLLM agents and tool use → Multimodal and embodied agents
  97. LOCI: Spatial Linear Memory for Streaming World ModelsWorld models → Video-generation world models and simulators +1
  98. PhantomEnvironments: Training LLM Agents in Fictional WorldsLLM agents and tool use → Reinforcement learning for LLM agents
  99. Learning Skills from Historical Action Trajectories: Action Experience Dictionary for World Action ModelsWorld models → Embodied and robot world models
  100. Social-WM: Safety-Aware Latent World Models for Robot Social NavigationWorld models → World models for navigation
All 937
  1. Learning from Research: Toward Lifelong Agent Harness EvolutionLLM agents and tool use → General LLM agent frameworks and benchmarks
  2. Reinforcement Learning-Guided Graph Transformations for SpTRSV OptimizationDeep reinforcement learning methods → RL for energy, networks and operations
  3. Dream4ACT: A Shared Visual Action Interface for Multi-Embodiment Video-Action ModelingDiffusion models → Video generation
  4. Role-Adaptive Policy Optimization for Offline Reinforcement LearningOffline RL and imitation learning from data → Offline RL algorithms
  5. Tactile Curiosity Drives Robot InteractionDeep reinforcement learning methods → RL for robot locomotion and manipulation
  6. Persistent Context Graphs for Efficient Memory Compaction in LLM AgentsLLM agents and tool use → Agent memory systems +1
  7. Agent Error Dataset: Scaling 50,000 Error--Diagnosis Pairs for Failure Analysis and Error-Aware Post-TrainingLLM agents and tool use → Software engineering coding agents
  8. JuryFlow: Disagreement-Guided Human-in-the-Loop Multi-Agent EvaluationLLM agents and tool use → Social simulation and multi-agent debate
  9. Efficient Expert-Parallel Communication on PCIe-Connected Consumer GPUsMixture-of-experts → Efficient MoE training and inference systems
  10. MeanVoiceFlow2: Joint Optimization of Mean Flow and Content Encoder for Fast One-Step Zero-Shot Voice ConversionDiffusion models → Audio, speech and music
  11. LongEmo: Towards Emotion Understanding and Reasoning in Long VideosLLM reasoning → Audio-language model reasoning +1
  12. Conversational Capture: A Trajectory-Level Framework for Evaluating Generative Engine Optimization in Multi-turn Human-Agent InteractionRetrieval-augmented generation → Agentic RAG, agents and memory +1
  13. Less Data, Better Timing: Student-Curriculum Coupling for VLM On-Policy Distillation in Temporal Video GroundingMultimodal LLMs / vision-language models → Video understanding with MLLMs +1
  14. CoEvoWhen: Policy-Tool Coevolution for Ultra-Long Video Temporal GroundingLLM agents and tool use → Multimodal and embodied agents
  15. Gromov-Wasserstein Distillation for Inductive Multi-View EmbeddingKnowledge distillation → General knowledge distillation methods
  16. Multi-Link Safety Filtering for VLA Policies Around Moving HazardsVision-language-action models → Safety and adversarial attacks
  17. DashVMC: Real-Time Discrete World Model Control in Geometry DashWorld models → Latent dynamics models and model-based RL
  18. Learning to Explain While Planning: Rule-Aligned Diffusion Planning for Autonomous DrivingEnd-to-end autonomous driving → Imitation-learning end-to-end planners
  19. What Can Component-Replacement Evidence Establish? A Critical Scoping Review of Local Decisions in LLM AgentsLLM agents and tool use → Tool learning and function calling
  20. Mid-Harness: Scaling Actions Between Model and Harness for Terminal AgentsLLM agents and tool use → Software engineering coding agents
  21. EWAM: Emergent Depth-Wise Specialization in a Unified Embodied Model -- From Semantic Understanding through Visual Foresight to ActionWorld models → Embodied and robot world models +1
  22. UBTree: Parallel Tree Drafting via Unigram and Bigram Models for Speculative DecodingEfficient LLM inference and serving → Speculative decoding
  23. When Instructions Retrieve Trajectories: Diagnosing and Mitigating Generalization Failures in VLA ModelsVision-language-action models → Failure recovery and robustness
  24. What Limits Recursive Reasoning Models: Optimization, Architecture and Test-Time ScalingLLM reasoning → Test-time compute and efficient thinking
  25. AIMS: An Agentic AI Framework for Sim-to-Real Multi-Modal ISACLLM agents and tool use → Agentic AI for wireless networks
  26. Better Deck or Different Judge? Evaluating Agentic Harness Gains in Corporate and Investment BankingLLM agents and tool use → Financial trading and forecasting agents
  27. Learning When and How to Intervene: A Hindsight-Distilled Sentinel for Coding AgentsLLM agents and tool use → Software engineering coding agents
  28. Learning to Reason with Compressed Context: Ground-Truth-Free Adaptation of OmniLLMs via Self-DistillationLLM reasoning → Audio-language model reasoning +1
  29. CoVisco: Codec-Native Vision Encoder with Native Token Compression for Unified Image-Video UnderstandingMultimodal LLMs / vision-language models → Visual token pruning and efficiency +1
  30. MCD: Causal Distillation of Multimodal In-Context Learning in Large Vision-Language ModelsMultimodal LLMs / vision-language models → MLLM training and visual instruction tuning +1
  31. NavHarness: Adaptive Goals for Agentic Vision-Language NavigationLLM agents and tool use → Multimodal and embodied agents
  32. The Concrete-Arbitrary Gap: Kinship Reasoning in LLMs Is Not Indifferent to PresentationLLM reasoning → Reasoning benchmarks and evaluation
  33. Patient-Centered Treatment Planning for Chronic Multimorbidity: A Hierarchical Reinforcement Learning Framework for Preference ModelingDeep reinforcement learning methods → RL for clinical treatment and surgery
  34. OSWorld-Science: A Benchmark of Computer Use Agents for Learning and Using Scientific SoftwareMultimodal LLMs / vision-language models → Chart understanding and anomaly detection +1
  35. CodeMimicry: Exploiting Safety Generalization Lag in Large Language Models via Structured Code CompletionLLM alignment / RLHF / preference optimisation → Safety alignment and jailbreak attacks +1
  36. Do Better Goal Representations Improve Goal-Conditioned Reinforcement Learning?Deep reinforcement learning methods → Deep RL algorithms and representations
  37. Learning Where to Look: Anatomical Grounding and Guided Attention for Cardiac MRI Vision-Language ModelsMultimodal LLMs / vision-language models → Medical vision-language models
  38. LLM Persona UnlearningLLM safety and security → Machine unlearning
  39. SplineWAM: Adaptive Action Horizons for World Action Models via B-Spline RepresentationsWorld models → Embodied and robot world models
  40. Hyperspectral Image Models: Technical ReportState-space sequence models (S4, Mamba) → Hyperspectral image analysis with Mamba
  41. Magic-W0: A Structured World-Action Foundation Model for Physical IntelligenceWorld models → Embodied and robot world models
  42. GrammarRL: Effective Grammar-Constrained Decoding via Reinforcement LearningDeep reinforcement learning methods → RL training of language and agent models
  43. Preemptive LLM Unlearning against Forbidden Capability Acquisition via Gradient SealingLLM safety and security → Prompt injection and agent security
  44. Fork-dLLM: Avoiding the Flexibility Trap in Diffusion Language ModelsDiffusion models → Discrete and text diffusion
  45. RL-Guided PAC-NMPC for Probabilistically-Safe Perception-Based Navigation in Unknown EnvironmentsDeep reinforcement learning methods → RL for robot navigation and UAV control
  46. Cognitive Enhancement: Rethinking the Necessity of Role-Playing for Large Language ModelsLLM reasoning → Reasoning benchmarks and evaluation
  47. SEAR: Spoofing Evidence-Grounded Audio Reasoning Benchmark for Audio Language ModelsLLM reasoning → Audio-language model reasoning
  48. When a Kindergartener Solves Calculus: Measuring Capability Leakage in Role-Prompted Reasoning ModelsLLM reasoning → Prompting and chain-of-thought reasoning
  49. Spherical Interpolation for Backward-Compatible Multimodal RepresentationsMultimodal LLMs / vision-language models → Document understanding and multimodal RAG
  50. KUAISHOU Explorer LLM-Rec Challenge 2026: Reasoning Generative RecommendationLLM reasoning → LLM reasoning for recommendation
  51. Toward Real-Time VLAs: Stage-Aware Two-Step Flow Denoising and System-Level EvaluationVision-language-action models → Efficient VLA inference and adaptation
  52. Learning from Runtime Feedback through Failure-Bank Self-Evolution for Vision-Language-Action ModelsVision-language-action models → Failure recovery and robustness
  53. RATIO: Reasoning Analysis and Token-level Inference Optimization for Quantized Reasoning ModelsLLM reasoning → Test-time compute and efficient thinking
  54. Inline Memory Meets Reusable Skills: Memory-centric Framework for Vision-Language-Action ModelVision-language-action models → Reasoning and chain-of-thought in VLA
  55. Explore-on-Graph: Hybrid Embedding-LLM Reasoning for Knowledge Graph Question Answering under IncompletenessLLM reasoning → Knowledge- and tool-augmented reasoning
  56. GraphMAS: A Systematic Benchmark of Multi-Agent Coordination for Graph LearningLLM agents and tool use → General LLM agent frameworks and benchmarks
  57. Security Properties of Neural Networks as Decision ProblemsLLM safety and security → Backdoor and data poisoning attacks
  58. DiffWAM: A Fast and Efficient Navigation World Action ModelWorld models → World models for navigation
  59. Revisiting On-policy Adversarial Black-Box Distillation: Calibrating Groupwise Reward Geometry for Effective Advantage ConstructionKnowledge distillation → On-policy and self-distillation for LLMs
  60. Beyond Policy Alignment: Closing the Planning-Learning Loop for Robot Control with Learned World ModelsWorld models → Latent dynamics models and model-based RL
  61. LatentHarness: Learning Latent Actions for Memory and Reasoning via Counterfactual Policy DistillationLLM reasoning → Test-time compute and efficient thinking
  62. OverForge: Reasoning Through Strategies and Tactics Helps Cooperative Lifelong AdaptationLLM reasoning → Reasoning for agents, planning and code +1
  63. Trust Is Not a Score: Runtime Assurance Contracts for High-Risk AI AgentsLLM agents and tool use → Agent threats, guardrails and governance
  64. BTC3D: Blended Tile Conditioning for Detail-Enhancing Image-to-3D GenerationDiffusion models → 3D, point cloud and scene generation
  65. When Masking Helps or Hurts Robustness in Compressed CLIP: A Pre-Deployment DiagnosticEfficient LLM inference and serving → Visual token pruning in multimodal models
  66. GFD-OPD: Guidance-Folded On-Policy Distillation of Diffusion Models Across ScalesDiffusion models → Discrete and text diffusion
  67. ShieldCLIP: Selective Safety Alignment for Harmful Content Mitigation in Multimodal Foundation ModelsLLM alignment / RLHF / preference optimisation → Safety alignment and jailbreak attacks +1
  68. Better Supervision Is Nearby: Neighborhood On-Policy Self-DistillationLLM reasoning → Reinforcement learning for reasoning +1
  69. RoboCoach: World Models as Active Coaches for Compositional Robot SkillsWorld models → Embodied and robot world models
  70. From Local Whole-Body VLA Behaviors to Scene-Scale Aerial ManipulationVision-language-action models → Aerial and drone VLA
  71. ChronoGraph: Functional 4D Scene Graphs with Vision-Language Models for Interaction Understanding and Grounded PlanningMultimodal LLMs / vision-language models → Embodied and 3D vision-language models
  72. Typographic Attack Against VLM-based AI-generated Image DetectionMultimodal LLMs / vision-language models → MLLM jailbreaks and adversarial robustness
  73. Diffusable Latents from Structure-Agnostic DistillationDiffusion models → Image generation, text-to-image and editing
  74. From Modes to Memories: Characterizing the Scale-Space Dynamics of Diffusion ModelsDiffusion models → Diffusion foundations and sampling
  75. Towards Better Exploration in Sequential Test-Time ScalingLLM reasoning → Test-time compute and efficient thinking
  76. MIND: Marginal-Invariant Neural Dependency Diffusion for Mixed-Type Tabular GenerationDiffusion models → Diffusion foundations and sampling
  77. D-Scope: Decomposing and Steering Diffusion Transformers with Sparse AutoencodersDiffusion models → Image generation, text-to-image and editing +1
  78. Pretext: Defeating Malicious Skill Detection Frameworks for AI AgentsLLM agents and tool use → Agent skill learning and evolution
  79. FOMO: Forget the Concept, Don't Miss Out on the Scene in Selective Video UnlearningLLM safety and security → Vision-language and text-to-image safety
  80. Why Do Conventional World Models Fail to Learn Cellular Automata?World models → Latent predictive world models +1
  81. GroundAnything: Reconciling Parallel Decoding with Precise Visual Grounding at Flash SpeedEfficient LLM inference and serving → Fast autoregressive visual generation
  82. KilometerVision: A New Frontier for Large-Scale Spatial Intelligence in VLMsMultimodal LLMs / vision-language models → Embodied and 3D vision-language models
  83. Thinking Outside the Box: Can Language Models Rely on External Guidance Selectively?LLM reasoning → Reasoning for agents, planning and code
  84. ECHO-G: Embodied Co-speech Humanoid mOtion GenerationDiffusion models → Audio, speech and music
  85. Steering Fields: Adaptive Vector Fields for Safe Image Generation and BeyondLLM safety and security → Vision-language and text-to-image safety +1
  86. From Given to Gathered Evidence: Agentic Learning for Longitudinal Medical ReasoningLLM reasoning → Clinical and biomedical applications +1
  87. A2Z GameSpec-Bench: How Faithfully Can Coding Agents Generate Games from Game Design Specifications?LLM agents and tool use → Software engineering coding agents
  88. Self-Repulsive Sampling for Diffusion Language ModelsDiffusion models → Discrete and text diffusion
  89. RankEvolve: A Reliable Multi-Agent Auto-Research Harness for Evolving Ranking ModelsLLM agents and tool use → Web, search, and research agents
  90. Speculative Safety Honeypot: Toward Proactive Defense Against Multi-turn Agent AttacksLLM agents and tool use → Agent threats, guardrails and governance
  91. Learning Normal Diffusion Dynamics for Backdoor Defense in Text-to-Image ModelsDiffusion models → Robustness, privacy and adversarial aspects +1
  92. Learning Reliable GUI Agents under Imperfect PriorsMultimodal LLMs / vision-language models → Embodied and 3D vision-language models +1
  93. Comparative study of adapting pre-trained models for driving behavior video captioningEnd-to-end autonomous driving → Large language models for driving
  94. CATCH: A Controllable Analysis Testbed for Reward Hacking in Coding RLLLM alignment / RLHF / preference optimisation → Reward hacking and misspecification
  95. WEIRDO: WEak resIdual Regularized DOob's h-transform diffusion alignmentDiffusion models → Diffusion foundations and sampling
  96. Discrete Forcing: Infusing Discrete Guidance into Continuous Denoising for Few-Step Action ExpertsVision-language-action models → Efficient VLA inference and adaptation +1
  97. Spike-driven Vision-Language-Action ModelVision-language-action models → Efficient VLA inference and adaptation
  98. LIBERO-Agent: Evaluating General-Purpose Agents for Direct Embodied ManipulationLLM agents and tool use → Multimodal and embodied agents
  99. PartiCam: Camera Controlled Video Generation with Reward GuidanceDiffusion models → Video generation
  100. Disentangling Self-Distillation: Measuring and Modeling Acquisition and RetentionKnowledge distillation → On-policy and self-distillation for LLMs
  101. Front-to-Back: Benchmarking Vision-Language Models for Asymmetric Cross-View Vehicle Re-IdentificationMultimodal LLMs / vision-language models → Driving VLMs and open-vocabulary detection
  102. OmniReasoning: Pushing the Limits of Audio-Visual Joint ReasoningKnowledge distillation → Multimodal distillation
  103. Correcting CondOT: Exact Finite-Step Sampling in Gaussian Flow MatchingDiffusion models → Flow matching, time series and science
  104. Also Small Models Can Reasonably Self-Evaluate Their ConfidenceLLM reasoning → Reasoning benchmarks and evaluation
  105. ActionGuard: Tool Call Authorization under Poisoned SkillsLLM agents and tool use → Agent threats, guardrails and governance
  106. Synthetic Data Characterization via Training DynamicsSynthetic data and self-improvement for LLMs → LLM synthetic data generation
  107. Raw-Routed Mixture of Adapters: A Causal Intervention for Routing Collapse in Time Series Foundation ModelsMixture-of-experts → MoE for time series and graphs
  108. CAST: Causal Advantage-Structured Training with Spatially Grounded Compositional Rewards for Diffusion ModelsDiffusion models → Diffusion foundations and sampling
  109. Network-based Spatial Context Retrieval for Open-weight LLMs: A Faithfulness Benchmark for Grounded Geographic ReasoningLLM reasoning → Spatial and embodied reasoning
  110. From Imitation to Reward Discovery: On-Policy Warmup for Agentic RLLLM agents and tool use → Reinforcement learning for LLM agents +1
  111. QuantCode Model: Specializing Language Models for Executable Algorithmic Trading CodeLLM agents and tool use → Financial trading and forecasting agents
  112. No Task Vector Is an Island: A Comprehensive Study on the Composability of Task Vectors from On-Policy DistillationKnowledge distillation → On-policy and self-distillation for LLMs
  113. Advancing Entropy-Level Credit Assignment in RLVR via Proximal Entropy Policy OptimizationLLM reasoning → Reinforcement learning for reasoning
  114. Experimental Experience Modeling for Autonomous ResearchLLM agents and tool use → Web, search, and research agents
  115. Decoupled and Distilled: Task-Adaptive LoRA-Teachers with Ensemble Knowledge Transfer for Few-Shot Class-Incremental LearningKnowledge distillation → Distillation for continual learning
  116. UniWAM Technical Report: Unified Mobile Manipulation via Mixed-Stream World-Action Modeling and Manipulation Anchor Pose SupervisionWorld models → Embodied and robot world models
  117. SkillFM: Generating Skills for LLM Agents via Latent Flow MatchingLLM agents and tool use → Agent skill learning and evolution
  118. Wavelet Flow Matching for Time SeriesDiffusion models → Flow matching, time series and science
  119. EHR-RobustGym: Benchmarking and Training Agents for Robust Clinical ReasoningLLM reasoning → Clinical and biomedical applications +1
  120. Rethinking Multi-Image Re-Representation in Multi-Image UnderstandingLLM reasoning → Multimodal reasoning
  121. Hiding in Plain Sight: Decoupling Pretext from Actuation for Skill Poisoning in LLM AgentsLLM agents and tool use → Agent threats, guardrails and governance
  122. HAPMoE: Heterogeneity-Aware Automatic Parallelism Planning for Mixture-of-Experts Models TrainingMixture-of-experts → Efficient MoE training and inference systems
  123. Offline Guidance, Online Reasoning: Reusing LLM Feedback for Small Language ModelsLLM reasoning → Reasoning benchmarks and evaluation
  124. Belief-Based Maximum Occupancy Principle and Active InferenceDeep reinforcement learning methods → Deep RL algorithms and representations
  125. Learning Beyond Full Imitation: Task-Preserving Knowledge DistillationKnowledge distillation → On-policy and self-distillation for LLMs
  126. WinoTS: Wavelet-based Self-Distillation for Time Series ModelsKnowledge distillation → Detection and dense prediction distillation
  127. Taming Speculative Search for Test-Time Scaling in LLM ServingLLM reasoning → Test-time compute and efficient thinking +1
  128. PatchKV: Weight-Space Compensation of KV CacheEfficient LLM inference and serving → KV cache optimization
  129. WorkGenesis: Building the Worlds That Teach Agents to WorkLLM agents and tool use → Agent skill learning and evolution
  130. MotionWeave: Learning Motion-Centered Future Dynamics for Vision-Language-Action PoliciesWorld models → Embodied and robot world models +1
  131. GRPO Training Dynamics for Small Language ModelsLLM reasoning → Reinforcement learning for reasoning
  132. ReSAIL: Mitigating Collapse in Iterative Agent Self-DistillationLLM agents and tool use → Agent skill learning and evolution +1
  133. Scale and Selection: What Makes Automatic Harness Evolution Work for Visual-Interface Robot AgentsLLM agents and tool use → Multimodal and embodied agents
  134. MiniRep: Robust Reputation-Based Aggregation for Multi-Agent DebateLLM agents and tool use → Agent threats, guardrails and governance
  135. ANI: Adaptive Numerical Injection for Unifying Semantic and Arithmetic Representations in Numerical ReasoningLLM reasoning → Mathematical reasoning
  136. Faithful Dual-constrained Erasure for Robust LLM Safety AlignmentLLM alignment / RLHF / preference optimisation → Safety alignment and jailbreak attacks +1
  137. ReTaCo: Residual-Target Control for On-Policy DistillationKnowledge distillation → On-policy and self-distillation for LLMs
  138. Concept Subspaces Compute Beyond the Logit Lens: A Weights-Only Test for Locating Representations Upstream of ReadoutMechanistic interpretability → Steering and interpreting LLM reasoning
  139. ReWAM: Reciprocal World Action Models for Interactive Autonomous DrivingWorld models → World models for autonomous driving +1
  140. Right Answer, Wrong Mechanism: Detecting Pernicious Divergence in Causal InterventionsMechanistic interpretability → Probing deception, bias and beliefs
  141. The Planning Limits of Latent World ModelsWorld models → Embodied and robot world models
  142. Emergent Multi-View Geometry Through Self-DistillationKnowledge distillation → Detection and dense prediction distillation
  143. DC-SAE: Deep Compression Semantic Autoencoder for Faster Diffusion ConvergenceDiffusion models → Image generation, text-to-image and editing
  144. Learning Process Rewards via Reasoning State PropagationLLM reasoning → Reinforcement learning for reasoning +1
  145. ASENA: Self-evolving Agents for Embodied NavigationLLM agents and tool use → Multimodal and embodied agents
  146. UniAE-MoE: A Unified Audio Encoder via Mixture of ExpertsMixture-of-experts → Multimodal MoE models
  147. DSDyn-VLA: A Dual-Stream Dynamic Manipulation Framework with Motion Perception, Future Awareness, and Realtime CorrectionVision-language-action models → Efficient VLA inference and adaptation
  148. Low-Discrepancy Dither for Quantized Recurrent State CachesState-space sequence models (S4, Mamba) → Core SSM architectures for language
  149. Aligning Thoughts with Answers: Probability Rewards to Tame Thinking DriftLLM reasoning → Multimodal reasoning +1
  150. MEND: Label-Free Detection, Localisation, and Correction of Latent Hallucination in World ModelsWorld models → Latent predictive world models
  151. LocoWM: High-Precision Locomotion through World-Model-Guided Residual AdaptationWorld models → Humanoid and legged locomotion world models
  152. Exploiting Vulnerabilities: Universal Adversarial Attacks on Vision-Language-Action Models in RoboticsVision-language-action models → Safety and adversarial attacks
  153. Whitening Improves Robustness to Spurious Correlations in Linear ProbesMechanistic interpretability → Sparse autoencoder training and evaluation
  154. Reinforcing Multimodal Reasoning via Token-Level Perception-Grounded Advantage EstimationLLM reasoning → Multimodal reasoning +1
  155. DAGent: Evaluate-then-Grow Planning for Deep Research AgentsLLM agents and tool use → Web, search, and research agents
  156. Linear Recurrent Memory Suffices to Distil a World-Model Policy for Robot Air HockeyWorld models → Latent dynamics models and model-based RL
  157. OP-CAD: On-Policy Clean-Audio Distillation for Robust Audio-Visual ReasoningLLM reasoning → Audio-language model reasoning +1
  158. Rep2Skill: Representation-Guided Skill Self-Evolution for LLM AgentsLLM agents and tool use → Agent skill learning and evolution
  159. MADBench: Benchmarking the Security of Multi-Agent DebateLLM reasoning → Reasoning for agents, planning and code +1
  160. Blackout vs. Freeze: Analyzing Physical Failure Modes of VLAs under Camera FaultsVision-language-action models → Failure recovery and robustness
  161. When Can Text Replace Vision? Structural Bottlenecks in Diagram ReasoningLLM reasoning → Multimodal reasoning
  162. Schema: Discovering Unknown Environments via Agentic Program InductionLLM agents and tool use → General LLM agent frameworks and benchmarks
  163. ID Balancing: Stable Training of Extremely Sparse MoE via PID-Based Load ControlMixture-of-experts → MoE large language models
  164. Asking the World: Generalist Physical Reasoning through Agentic World Modeling and ProbingWorld models → World models for language agents +1
  165. Feature-Aware Token Attack for Compression-Triggered Stealthy Failures in Large Vision-Language ModelsMultimodal LLMs / vision-language models → Visual token pruning and efficiency +1
  166. Characterizing High Bandwidth Flash for LLM ServingEfficient LLM inference and serving → Serving systems and request scheduling
  167. CDMD: A Cross-Dataset Mixed-Type Diffusion Model for Tabular DataDiffusion models → Diffusion foundations and sampling
  168. Is Better Teacher Supervision Enough? Unlocking Student-side Learning in Multimodal On-Policy DistillationKnowledge distillation → On-policy and self-distillation for LLMs
  169. Diagnosing On-Policy Self-Distillation for Reasoning Language ModelsLLM reasoning → Prompting and chain-of-thought reasoning +1
  170. CamAgent: An LLM-Agent Framework for Multi-Species Camera-Trap WorkflowsLLM agents and tool use → Multimodal and embodied agents
  171. MASCRDM: Multi-Agent System for Compliance Risk Detection and Mitigation in Training Process of Large Language ModelsLLM agents and tool use → Agent threats, guardrails and governance
  172. Beyond Prediction: Steering VLM Agents with Retrospective World ModelingWorld models → Latent dynamics models and model-based RL +1
  173. DiFF: Doppler-informed Flow Matching for Human Motion FlowDiffusion models → Diffusion and flow policies for robotics
  174. DeCoPrune: Efficient KV-Cache Pruning for Autoregressive Video Diffusion via Denoising ConsistencyDiffusion models → Video generation +1
  175. Learning Infinite-Horizon Average-Reward CMDPs via State AugmentationDeep reinforcement learning methods → RL theory and sample complexity
  176. Coding Agents for Coding TheoryLLM agents and tool use → Software engineering coding agents
  177. Multi-LLM Collaborative Alignment via Stackelberg GamesLLM alignment / RLHF / preference optimisation → LLM preference optimization methods
  178. RAGScope: A Leakage-Controlled, Cost-Aware Evidence-Gating Protocol for RAG Hallucination TriageRetrieval-augmented generation → Hallucination detection in RAG
  179. LexReward: A Taxonomy-Driven Reward Framework for Legal Language ModelsLLM alignment / RLHF / preference optimisation → Process reward models for reasoning
  180. CORE: Conflict-Oriented Reasoning Elimination for Verifiable Language-Model SearchLLM reasoning → Test-time compute and efficient thinking
  181. SparseEngine: Sparse-First Inference EngineLLM agents and tool use → Agent memory systems +1
  182. Agentic Tool-Augmented Reasoning for Explainable Image Forgery DetectionLLM agents and tool use → Multimodal and embodied agents
  183. Can Agents Trust Their Skills? Uncovering Unsafe Chains of Trust in Skill-Based LLM AgentsLLM agents and tool use → Agent threats, guardrails and governance
  184. SteerQuant: Steering Quantization Error with Action-Guided Scaling in World-Action ModelsWorld models → Video-generation world models and simulators
  185. Covert Assistance: Helpful LLM Agents Evade Oversight in Multi-Agent SystemsLLM agents and tool use → Agent threats, guardrails and governance
  186. BadAction: Backdoor Attacks on Interactive Video Generation via Action-Guided TriggersLLM safety and security → Backdoor and data poisoning attacks
  187. RSIGame: Autonomous Agentic Game Development with Recursive Self-improvementLLM agents and tool use → Reinforcement learning for LLM agents +1
  188. Looking Back to Move Forward: Temporal Verification for Generative Robot PoliciesOffline RL and imitation learning from data → Language-conditioned policies and VLAs
  189. TED:Text-Axis Evidence Decomposition for Prompted Anomaly LocalizationMultimodal LLMs / vision-language models → Chart understanding and anomaly detection
  190. Search Shapes Conclusions: Auditing Evidence Selection Bias in Deep Research AgentsLLM agents and tool use → Web, search, and research agents
  191. Persistent Watermarking of Text-to-Image ModelsLLM safety and security → Watermarking generated content and models
  192. From Verification Failures to Reusable Guidance for Coding AgentsLLM agents and tool use → Software engineering coding agents
  193. Frame Differential On-Policy Self-Distillation for Video ReasoningLLM reasoning → Reinforcement learning for reasoning +2
  194. RouteRec: Behavior-Guided Sparse Routing for Sequential RecommendationMixture-of-experts → MoE for recommender systems
  195. When Clipping Reverses Correction: Failure Dynamics of Pointwise Forward-KL On-Policy Self-DistillationKnowledge distillation → On-policy and self-distillation for LLMs
  196. Cue the Flow: Steering Flow-Matching Policies for Open-World Delivery ManipulationDiffusion models → Diffusion and flow policies for robotics
  197. Smaller Models, Better Rejects: Preference Distillation ScalingKnowledge distillation → On-policy and self-distillation for LLMs
  198. Sparse-WAM: Accelerating World Action Models via Action-Guided Sparse ImaginationWorld models → Embodied and robot world models
  199. SimEX: Simulation-Integrated Robotics AutoResearchLLM agents and tool use → Software engineering coding agents
  200. Mitigating Object Hallucination in Large Vision-Language Models via False Discovery Controlled Visual Data SplittingMultimodal LLMs / vision-language models → Hallucination mitigation in LVLMs
  201. PARK: Accurate Block Retrieval for Sparse Attention in Video Diffusion TransformersDiffusion models → Video generation
  202. RealWorldShop: Benchmarking and Improving Conversational Shopping Agents in Real-World E-commerceLLM agents and tool use → Agentic recommender systems
  203. Making LLMs Say What They Think: Measuring and Improving CoT-Interpretability AlignmentLLM reasoning → Prompting and chain-of-thought reasoning
  204. Refusals That Bend: Measuring and Predicting Task Malleability in Embodied VLM PlannersMultimodal LLMs / vision-language models → Embodied and 3D vision-language models +1
  205. When Order Matters: First-Speaker Bias and Mitigation through Personality in Sequential Multi-Agent DebateLLM reasoning → Reasoning benchmarks and evaluation +1
  206. Alleviating Hallucination in Reasoning Tasks with Training-Free Uncertainty-Guided SteeringLLM reasoning → Hallucination detection and mitigation
  207. Targeted Retrieval, Compact Representations: How CoT Reasoning Improves Long-Context CountingLLM reasoning → Prompting and chain-of-thought reasoning
  208. Loop-Free Inverse Reinforcement Learning via Sequential Value Recovery with Q-Score MatchingDiffusion models → Diffusion and flow policies for robotics +2
  209. DrivingBench: Can Vision-Language Models Drive a Toyota Corolla?Multimodal LLMs / vision-language models → Driving VLMs and open-vocabulary detection
  210. Robust Risk-Sensitive Reinforcement Learning from Corrupted Human FeedbackLLM alignment / RLHF / preference optimisation → Reward modeling and RLHF +1
  211. Adaptive Self-Consistency: From Black-Box Sampling to Distribution-Valued FeedbackLLM reasoning → Test-time compute and efficient thinking
  212. Learning What to Forget: Distributional Unlearning for LLM Representation SpacesLLM safety and security → Machine unlearning
  213. World-as-Graph: Relational World Modeling Through Latent Space GraphsWorld models → Latent predictive world models
  214. From Image Interpretation to Clinical Reasoning: Upstream Physician-Context-Aware Multimodal Learning with Causal Reinforcement LearningLLM reasoning → Clinical and biomedical applications
  215. Flow Matching under Noisy Latent Structure: Beyond Exact Low-Dimensional SupportDiffusion models → Flow matching, time series and science
  216. Composing Task-specific Agent Harnesses at Test Time with Reusable PrimitivesLLM agents and tool use → Software engineering coding agents
  217. Unlearning Deceptive Behaviors in LLMs with Contrastive Forget SetsLLM safety and security → Machine unlearning
  218. DAMPER: Return-Prioritized Gradient Control for Smooth PoliciesDeep reinforcement learning methods → RL theory and sample complexity
  219. SceneJail: Exploiting Video Scenario Context to Jailbreak Multimodal LLMsMultimodal LLMs / vision-language models → MLLM jailbreaks and adversarial robustness +1
  220. K2P: Label-Free Knowledge to Prompt DistillationKnowledge distillation → Knowledge distillation for LLMs
  221. Unmerge: Efficient Machine Unlearning via Task ArithmeticLLM safety and security → Machine unlearning
  222. Consistent Plan-Act for Long-Horizon Agentic TasksLLM agents and tool use → General LLM agent frameworks and benchmarks
  223. PRICE the Action Chunks: Physical Relational Credit Assignment for Embodied Reinforcement LearningDeep reinforcement learning methods → RL for robot locomotion and manipulation
  224. STRATA: Self-Learning Through Role-Aligned Tiered Agents for Real-Time Strategy GamesLLM agents and tool use → Reinforcement learning for LLM agents
  225. Does Learning Protein Folding Generalize to Broader Reasoning?LLM reasoning → Knowledge- and tool-augmented reasoning
  226. DODGER: Safety-Guided Reinforcement Learning for Robot Navigation Among Dynamic ObstaclesDeep reinforcement learning methods → Safe and constrained RL
  227. Talk2Agent: Benchmarking Voice Interfaces for Text AgentsLLM agents and tool use → General LLM agent frameworks and benchmarks
  228. When Context Changes: Understanding Update Failures in LLMsLLM agents and tool use → Agent memory systems
  229. Efficient Multi-Modal Planning with Reward-Guided Preference Optimization for Autonomous DrivingEnd-to-end autonomous driving → Imitation-learning end-to-end planners +1
  230. Optimal Design for Active Preference Learning with Biased LLM JudgesLLM alignment / RLHF / preference optimisation → LLM preference optimization methods
  231. Online Evolution Strategy for Flow-Matching VLA Policies via Self-Supervised Trajectory Distribution OptimizationVision-language-action models → Reinforcement learning and post-training +1
  232. Mitigating the Length-Scaling Tax with Online DistillationLLM reasoning → Reinforcement learning for reasoning +1
  233. Visualizing Distribution Coverage in Generative Diffusion ModelsDiffusion models → Diffusion foundations and sampling
  234. Locomotion-Grounded Humanoid Soccer: Task-Gated Reinforcement Learning of a Multi-Directional Kicking LibraryDeep reinforcement learning methods → RL for robot locomotion and manipulation
  235. OpenJev-RLCD: A Working RLCD ImplementationLLM reasoning → Reinforcement learning for reasoning
  236. Scoring Higher, Answering Worse: Mitigating Reward Hacking in Rubric-Based RL via Protocol-Level RubricsLLM alignment / RLHF / preference optimisation → Reward hacking and misspecification +1
  237. SparLeak: Privacy Leakage from Sparse Attention in LLM Inference on Shared GPUsEfficient LLM inference and serving → KV cache optimization
  238. Diversity Combining for Multi-Path LLM ReasoningLLM reasoning → Test-time compute and efficient thinking
  239. DecoMoE: Decoupling Visual Propagation and Expert Computation for Efficient Multimodal MoE InferenceMixture-of-experts → Vision mixture-of-experts
  240. SkillSeek: Revisiting Agent Skill Retrieval at Marketplace ScaleLLM agents and tool use → Agent skill learning and evolution
  241. Future Video Generation Better Aligns with the Human Visual Cortex than Observed VideoDiffusion models → Video generation
  242. When Reasoning Goes Astray: Attention Dynamics of Uncontrolled ReasoningLLM reasoning → Test-time compute and efficient thinking
  243. CRAFT: Causal Responsibility and Failure Tracing in Medical Vision Language ModelsMultimodal LLMs / vision-language models → Medical vision-language models
  244. StateTree: Enhancing Long-Term Dialogue Reasoning via Reinforcement LearningLLM reasoning → Test-time compute and efficient thinking
  245. PatchHolmes: Agentic Patch Retrieval via Listwise SelectionLLM agents and tool use → Software engineering coding agents
  246. Blackboard Intelligence Can Surpass Autoregressive on Globally Constrained ProblemsDiffusion models → Discrete and text diffusion
  247. Explicit Trajectory Diversity for RL-Based Post-Training of LLM AgentsLLM agents and tool use → Reinforcement learning for LLM agents
  248. GraphCert: Bootstrap Agentic Graph Reasoning with Certified Evidence RubricsLLM reasoning → Knowledge- and tool-augmented reasoning +2
  249. Recovering Off-Policy Supervision for Speculative DecodingEfficient LLM inference and serving → Speculative decoding
  250. Training LLM Judges from Language Feedback via Position-Selective Self-DistillationKnowledge distillation → On-policy and self-distillation for LLMs
  251. Persona and Persuasive Framing in AI Voice Agents: A 2×2 Field Experiment with ChildrenLLM agents and tool use → Social simulation and multi-agent debate
  252. ChartDensity-Bench: Benchmarking MLLMs for Numerical Data Reconstruction under Visual DensityMultimodal LLMs / vision-language models → Chart understanding and anomaly detection
  253. Action Conditioned Bisimulation For GUI Agent MemoryLLM agents and tool use → Agent memory systems
  254. Distill the Visual Evidence, Not Just the Answer: Cross-World On-Policy Distillation for Vision-Language ModelsMultimodal LLMs / vision-language models → VLM benchmarks and visual reasoning +1
  255. Distilling Diffusion Score Discrepancy for Efficient Training Data AttributionDiffusion models → Diffusion foundations and sampling
  256. Self-Evolving Algorithm-Design Agents: Escaping In-Context Evolutionary Stagnation via Population-Curated Policy OptimizationLLM agents and tool use → General LLM agent frameworks and benchmarks +1
  257. Learning to Route in Visual Space via Multi-Step Embedding RetrievalLLM agents and tool use → Multimodal and embodied agents
  258. Code to Control: Synthesizing Parameterized Reactive ControllersWorld models → Latent dynamics models and model-based RL
  259. Anchor-ECC: Local Integrity Checking for Watermarked LLM Outputs via Error-Correcting CodesLLM safety and security → Watermarking generated content and models
  260. UniEvo-VL: An On-policy Self-Distillation Training Recipe for Multimodal Model Self-improvementSynthetic data and self-improvement for LLMs → Self-improving MLLMs and synthetic data +1
  261. Soft Spatial ReasoningLLM reasoning → Spatial and embodied reasoning +1
  262. SpatialCORE: Confidence-Aware Grounded Spatial Reasoning in Large Vision--Language ModelsLLM reasoning → Spatial and embodied reasoning +1
  263. Staying on Task: Testing the Foundations of Long-Horizon Agent ReliabilityLLM agents and tool use → Software engineering coding agents
  264. Budget Boundary Effects in Test-Time Mathematical ReasoningLLM reasoning → Test-time compute and efficient thinking
  265. Cascadia: A Control-Plane-Free Alternative to Hyperconverged AI InfrastructureEfficient LLM inference and serving → Serving systems and request scheduling
  266. ReGain: Restoring Subject Fidelity in Personalization on Synthetic ImagesDiffusion models → Image generation, text-to-image and editing
  267. Context Injection and Long-Horizon Reliability: Separating Duplicate Recognition, Local Processing, and Counter Tracking
  268. Real-time Chatbot Romance Scam Detection in Stage Classification
  269. PLDiv: Measuring Dataset Diversity and Semantic Collapse via Persistence Landscapes
  270. Rotate the Loop: Blockwise Conformality for Looped Transformers
  271. Can We Trust This Evaluation? Instance-Level Reliability of LLM-as-a-Judge
  272. Pick the Mixture, Not the Samples: Combinatorial Data Selection for LLMs
  273. EarLog: Opportunistic Ear Physiology Sensing toward Personalized Hearing Risk Awareness
  274. Acoustic Biomarkers for Alzheimer's Disease Detection from Spontaneous Speech
  275. Beyond Risk Prediction: Evidence Grounding and Psychosocial Factor Verification for Explainable Suicide Risk Assessment
  276. Parameter-Free Interval-Dynamic Regret under Heavy-Tailed Noise
  277. TRACE: A Reproducible Benchmark for Electricity Price Forecasting with Official Operational Text
  278. MACTS-EM: Multi-Agent Collaborative Time Series Forecasting with Emergent Memory
  279. Overcoming Challenges of Interpretive Structural Modeling with Large Language Models
  280. Approximation Property of Dropout Neural Networks: Sobolev Rates and Confidence Bounds
  281. Counterfactual Predictions in Scientific Emulators Without Controlled Experiments
  282. Nearest-neighbour baselines for fingerprint prediction from MS/MS spectra under different assumptions
  283. A Missing Latent, Not a Missing Simulator: Radius-Augmented Inference for Real JWST Retrieval
  284. Neural Fourier Surrogates for Data Reuploading Quantum Neural Networks
  285. Contextual trajectory and incremental contextual displacement: Towards using LLMs to understand dynamic, utterance-specific meaning construction
  286. TrueMuse: A Benchmark for Data Attribution in Text-to-Music Models
  287. VERITYGATE: A Four-Gate Schema-Level Faithfulness Framework and Paired Benchmark for Grounded LLM Narrations over Structured Evidence
  288. AnyJev Technical Report
  289. Verbalized and Internal Probabilities Are Coupled in Large Language Models
  290. On-the-fly Weight Generation: A Hypernetwork Proof of Concept on ARC-1D
  291. PI-AMFM: Permutation-Invariant Learning for Variable-Cardinality AM-FM Mode Decomposition in Biomedical Signal Analysis
  292. Quantifying the Impact of Ambulance Ramping: A Multi-Year Analysis of Victorian Emergency Medical Services Cases
  293. TabJoinBench: A Benchmark for Joinable Table Discovery
  294. Video Generation Models: A Survey of Post-Training and Alignment
  295. Spatially Gated Diffusion for Localized Counterfactual Chest Radiograph Editing
  296. Sapien: A Stateful Policy Engine for Autonomous AI Agents
  297. Can large language models unlock discrete data in ophthalmic diagnostic reports?
  298. Inference for stochastic differential equations driven by weighted sub-fractional Brownian motion using neural networks and the Euler approximation
  299. Enterprise Representation Simplification (ERS): Reducing Representational Complexity for Enterprise AI
  300. Localizing Transfer Between Memorization Tasks
  301. Pre-training interventions, ex post facto: Grafting model beliefs across checkpoints
  302. Crossing the Cyber Divide: Sim-to-Sim and Sim-to-Real Transfer for RL Agents
  303. Learning to Price Electricity for Optimal Demand Response
  304. Increasing Width Allows Greedy Layer-wise Training to Rival End-to-End Backpropagation in Self-Supervised Learning
  305. Signal-Noise Factorization Isolates Nuisance Variation into Removable Subspaces
  306. What Builds the Scene? Luminance Dominates Geometry Formation in 3D Gaussian Splatting
  307. StabilityArc: Decoding Protein Sequence Embeddings into Generalizable Stability Landscapes
  308. Q-MINO: A Minimal-Norm Method for Quantization-Aware Training
  309. Personalized Image Generation with Reasoning and Reflection
  310. Reformulation-Contrastive Learning for Mixed Integer Programs
  311. Reward as Observation: Learning Reward-Based Policies for Rapid Adaptation
  312. Outer Diversity of Condorcet Domains
  313. Sequential Functional Structured Tucker Compression for Large Language Model Attentions
  314. WOMBAT: Whitebox Oracle for Molecular Benchmarking and Attribution Testing
  315. Beyond Unimodal Bases: Pullback Geometry for Multimodal Data
  316. Initialization Improves LLM-Driven Discovery
  317. AnchorPrompt: Self-Distilled Soft Prompts for Robust Audio-Language Models
  318. R-GroundBench: A Diagnostic Benchmark for R-Group Groundingin Markush Molecular Editing
  319. Multifidelity Formulations for Triangular Transport
  320. Progressive-Resolution Secure Aggregation for Federated Learning
  321. How Divergence Becomes Decision Flips in Compressed Language Models
  322. FedMAD: Modulation-Aware Directional Aggregation for Federated Learning in Remote Sensing Image Classification
  323. Soundwich: Video Generation with Layered and Controllable Audio
  324. Towards Robust Numerical Claim Verification
  325. Leto: Fast In-Place Recovery for LLM Training on Surviving Hardware
  326. Grand Canonical Generators
  327. Curvature Under Attack in hZACH-ViT: Gauge Symmetry, Boundary Saturation, and Adversarial Failure
  328. Bayesian Fine-tuning Yields Language Models that are as Bayesian as their Beliefs Allow
  329. Nonparametric Distribution Matching for Self-Supervised Whole-Slide Image Condensation
  330. Harnessing Vision-Language Models for Perceptual Quality Assessment and Autonomous Content Adjustment in Augmented Reality
  331. Learning Transferable Skills using Goal-Conditioned Bisimulation
  332. LabBook: Harnessing Experimental History for Efficient LLM-Driven Discovery
  333. Closing the Loop: Practical Training Recipes for Looped Language Models
  334. ORBIT-FMIB: Tracking Order-Resolved Epistatic Information Through ESM-2
  335. A Simple Doxastic Deontic Logic for Norm-Guided Decision Making
  336. Analysis of Quantized and Efficiently Adapted Protein Language Models
  337. PhysicsMate: A Curriculum-Grounded Bengali Benchmark for Secondary Physics QA with Small-Model Adaptation
  338. Lingtai: What Concept Geometry Reveals--and Does Not Reveal--About LLM Inference
  339. When More Data Is Not Enough: The Context-Sufficiency Frontier in Generative AI Personalization
  340. On Evaluating Quantum Kernel Robustness for Low-Resource Cross-Corpus Audio Deepfake Detection
  341. Group-Invariant Statistics Determine Embedding Geometry: Harmonic Analysis of Representations from Bach to the Night Sky
  342. Learning Linear Systems under Heavy-Tailed Noise: A Non-Asymptotic Analysis from A Single Trajectory
  343. Mixture of Decoders for Diverse Dialog Response Generation
  344. Misalignment of Low-Loss Regions Causes Grokking
  345. Learning the identity: a case study of how SGD selects among functional decompositions
  346. Explainable Suicide Risk Assessment on Social Media with Multi-Task QLoRA
  347. End-to-End Historical Music Restoration in Latent Space
  348. Where's Waldo? Query-language Preference under Cross-lingual Knowledge Disparities
  349. Just Align bmx: Aligning Predictions, Not Representations
  350. Mean Spatial Frequency Decoupling for Learning-Based Uplink-to-Downlink Covariance Conversion in FDD Massive MIMO
  351. Towards Hierarchical Cyber Defense with Large Language Models: From Planning to Execution
  352. Right In-Place (RiP) Convolution: A Simple, General, and Near-Optimal Strategy for Memory-Efficient CNN Inference
  353. Worse Together: How Performance Breaks Down in Multi-User Multi-Agent Teams
  354. From Task Mixtures to Specialized Experts
  355. Query-efficient winner prediction in district-based elections
  356. Make Sparse Rewards Count: Density-Aware Reward Aggregation for Multi-Reward RL
  357. Scaling Collider Event Generation with Residual-Quantized Tokens
  358. Attention Kernels for Learning Maps Between Heavy-Tailed Measures
  359. Beyond Affine Transformations: A Soft Dominance Layer for Coordinate-Wise Neural Computation
  360. Evaluating Hybrid Quantum-Classical Models for Reduced-Order Brain Deformation Dynamics
  361. Generative Modeling of Stochastic Dynamics for Long-Time Evolution
  362. Geometry-Dependent Bounds for Online Non-Monotone DR-Submodular Maximization
  363. Memorizon: Training World Models Beyond Their Context Window
  364. Random Recursive Models
  365. Assessing the Impact of Language Disparity on Multilingual Linguistic Ability in Large Language Models
  366. Multi-agent Auditory Scene Analysis: Improved Localization Speed and Robustness by Multi-beamformed Speech Quality Feedback
  367. Adaptive Conformal Prediction for Image Regression Models with Application to an Inertial Confinement Fusion Emulator
  368. Science or Slop?: Benchmarking and Mitigating Scientific Slop in AI-Generated Papers
  369. Ontology-Based Contextual AI Evaluations (OB-CAIE) Methodology
  370. Rules Amortize, Pairings Don't: Linguistic Structure Determines What Latent Task Representations Can Replace In-Context Learning
  371. SimplexUQ: An Evaluation Framework and Benchmark for Conformal Uncertainty on Simplex-Valued Predictions
  372. One-Step Generative Modeling via Training Dynamics Action
  373. Fractional Laplace Neural Operators: Exact Architectures, an Expressivity Frontier at Criticality, and Certified Stability for Memory-Driven Network Dynamics
  374. Heteroskedastic Canonical Polyadic Tensor Decomposition
  375. Gumbel Straight Flow: Distilling Autoregressive Models into One-step Flow Maps
  376. ScaffoldM3C: A Multimodal Sequential Monte Carlo Framework for Generative Stable Construction Planning
  377. AIR-LLM: Broadcasting AI Weights over Radio for Memory-Free Edge LLM Inference via RF Computing
  378. PACT: End-to-End Learning of Human Pose, Contacts, and Forces from Video
  379. Frozen Scenes, Shifting Winners: Configuration Fragility in Text-to-3D Evaluation
  380. Exact information accounting for SGD methods
  381. One pool, many targets: a conservation layer and what archival data can identify
  382. Every Batch Is Its Own Validation Set: Leave-One-Out Gradient Matching for Online Data Selection in LLM Fine-Tuning
  383. XOR-Trellis: Ultra-Low-Complexity Dequantization and Curvature-Aware Hadamard-Free LLM Quantization
  384. ChainLoRA: Geometry-Preserving Task Vector Merging for Continual Learning in LLMs
  385. IrekoGPT: Turning Structured Pruning into Post-Hoc Slimmable LLMs
  386. Target-Dependent Limits of Causal Repair: A Leading-Log Frontier in a Gaussian Model
  387. The Life Cycle of a Massive Activation: Stochastic Birth, Weight-Decay-Driven Growth, and Competitive Consolidation
  388. Learning to Cover Locally: Graph Neural Combinatorial Optimization under a Hard Information Horizon
  389. Scores That Hold, Benchmarks That Leak: Measuring Dataset Contamination in Public Brain-Tumor MRI Classification
  390. Transferable Graph Metanetworks
  391. Benchmarking Prompt Optimization of Large Language Models With Chess
  392. Do Better Scores Mean Better Physics? Physics-Grounded Explanations for Sim2Real Neural Operators
  393. From Image Latent Space to Fuzzy Rules: Interpretable Analysis of Gastrointestinal Foundation Model
  394. VANDAM: Viewing a nucleotide sequence with DNA molecular priors
  395. UniBuc at SemEval-2024 Task 2: Tailored Prompting with Solar for Clinical NLI
  396. RACE: Residual-Aware Test-Time Adaptation for Neighbor-Rich Time-Series Foundation Model Forecasting
  397. The Conflict Between Logic and Memory: Training Conditions for Optimizer-Dependent Rule Acquisition
  398. Dissonant ballerinas and crafty carrots: a comparative multi-modal analysis of Italian brain rot
  399. Metacognitive Reasoning in Energy Based Models using Instance Based Learning Theory
  400. WIPSNet: Deep Learning for Paediatric Wheeze Detection from Overnight Impedance Pneumography
  401. Forking: Sudden Overfitting Under Replay
  402. Interpretable Synthetic Medical Tabular Data Generation for Clinical Decision Support Using Fuzzy Cognitive Maps
  403. MatrixReward: Reward from Rubric Matrix for Open-Ended Generation
  404. FAER: Auditable Utility-Aligned Trajectory Replay for Language Model Post-Training
  405. RIQE: a NIQE-style reference model for Computed Tomography
  406. EvoGen-Harness: Learning Where and How to Evolve Image-Generation Harnesses
  407. On the Relationship between Model Quantization and Model Inversion Attacks
  408. OmniMed-Jev: Calibrating LVLM Confidence for Trustworthy Medical Multimodal Decisions via System One
  409. Fusion techniques of time frequency-based images to predict the outcome of rTMS depression therapy
  410. STCFormer: Adaptive Spatio-Temporal Modeling with Dynamic Cluster Transformer for Station-based Weather Forecasting
  411. A First Glance at Jev for Network Traffic Classification: Accuracy, Processing Time, and Cost
  412. Faithful Chart Generation for Multimodal Deep Research: Frame-Evidence Co-Adaptation
  413. When Do Attention-Head Ablations Support Causal Claims? Projection-Level Confounds, Floor Effects, and Matched Controls
  414. Deny Without Disabling: Authorization-Paired Evaluation and Control for Multi-Agent Systems
  415. M2Weather: A Benchmark for Joint Multi-Station and Multi-Variable Weather Forecasting
  416. A Shared Taste for Model-Written Text: The Generator-by-Selector Matrices of "AI-AI Bias" Show No Detectable Own-Model Premium
  417. What Should an Agent Remember? Disentangling Retention from Retrieval in Bounded-Memory Evaluation
  418. Deep Learning for Anomaly Detection in Railway Systems: A Structured Survey
  419. DexPolicy: Scheduled Exploration for Trajectory-Guided Dexterous Manipulation
  420. Proof-Gated Signing: Solver-Checked Transaction Guards that Hold Under State Drift for Onchain AI Agents
  421. JusticeAxis: Benchmarking Legal Judgment between Rigid Rule Application and Ungrounded Discretion
  422. Multimodal Flow: Unified Flow Modeling of Language and Vision in Embedding Spaces
  423. Removing Timing Shortcuts Improves Non-Invasive Brain-to-Text
  424. ViTeX-Bench: Benchmarking High-Fidelity Video Scene Text Editing
  425. AssemblyWorld: Rethinking 3D Assembly with General-Purpose Agents
  426. Image Classifiers are Efficient Self-Supervised Video Representation Learners
  427. Learning Global Sensitivity Indices from Observational Data: A Metamodel-Based Approach
  428. Ego4WAM: What Matters When Scaling Egocentric Human Data for Robot Learning?
  429. EvoDuet: Bilevel Co-Evolution of Web Searching and Task Solving for Scientific Discovery
  430. Is Weight Tying Still Beneficial for Decoder-Only LLMs in Private Settings Under DP-SGD?
  431. I Have a Stream: Making Self-Supervised Learning Work on Continuous Video
  432. Turbo Harness: Instance-Adaptive Harness Optimization
  433. MatLoom: Layered Text-to-Material Generation in a Compact Program Space
  434. Atomizer-IO: Beyond Pixels, Patches and Grids
  435. GLARE: Generating Listening Heads with Appropriate Reactions
  436. Compression Footprints as Security Signals for Model-Poisoning Defense in Federated Learning
  437. DynaHarness: A Dynamic Physical Harness for Self-Evolving Robot Agents
  438. How Much Is an AI Token Worth? Scaling Laws for Wild AI-Generated Web Text
  439. Disentangling Computation in Multi-Task Neural Networks with the Green's Operator
  440. CAS II: Orbits as Models: Kolmogorov's Structure Function under Symmetry
  441. PMosFM: Preconditioned Manifold Matching for One-Step Physics-Constrained Generation
  442. Signal Processing over Product DAGs: Causal Shifts and Filters
  443. Belief-Aware Multi-Agent Path Finding under Map Uncertainty
  444. STARS: From Spatiotemporal Dynamics to Social Representations in Human-Robot Interaction
  445. StreamRig: Exploiting Intra-Rig Geometry for Streaming Multi-Camera Odometry
  446. Decision-Oriented Recommendation Reranking: An Empirical Study of Its Promise and Limitations
  447. Comparison of techniques for fine-tuning open-weight models for entity extraction from radiology reports
  448. Recognition of Urbanized Areas in UAV-Derived Very-High-Resolution Visible-Light Imagery
  449. SCB: SpeechConversationBench for Evaluating Multi-Turn Reasoning in Speech-to-Speech Models
  450. MemLife: Curating and Reasoning over Long-Term Egocentric Video Memories
  451. Near-Linear Accuracy Bounds for Moreau--Yosida Unadjusted Langevin Sampling
  452. Cheap to Draw, Expensive to Trust: Certifying Test-Time Scaling Curves
  453. Provably Tractable NFA-Constrained Language Generation via HMMs
  454. Index-Translate: A Multilingual Translation Model Family -- Text, Speech, Controlled Dubbing, and Long-Document Translation
  455. MANET-GNN: Learned Decentralized Optimization of Power Allocation in Multi-Channel MANETs
  456. PrefPI: Preference-Guided Steering into Out-of-Distribution Behaviors
  457. Partial identification with entropy regularized optimal transport
  458. From Spectra to Joint Schedules in LLM Pre-training: 3+3(+2) Scaling-Law Regimes
  459. Policy Iteration Is Not Strongly Polynomial for Deterministic Markov Decision Processes: The Price of Algorithmic Anarchy
  460. From DNA Design to DNA Slimming: Auditable Agentic Discovery of a Deletion-Only Designer
  461. Less is more: error-distance scaling relation for data-efficient kilometer-scale downscaling of extreme heat
  462. Game-Guided Skill Discovery through Self-Play for Playable Agent Control
  463. Prototype-Rule Neurosymbolic Regularization for Rank-Constrained Tensor Neural Networks under Label Scarcity
  464. VR-JEPA: Learning Contrastive-State Latent Guidance for Generation-based Video Reasoning
  465. Learning Functional Subspaces for Neural Network Compression
  466. Debias It Yourself: Teaching LLMs Cognitive Bias Mitigation Interventions
  467. On the (In)effectiveness of AMR Augmentation for Large Language Models
  468. Scalable Cox Regression via Grouped Risk Sets and Sharper LogSumExp Rates
  469. Beyond Model Ranking: Regime Diagnosis for Distributional-Statistical Misspecification in Industrial Time-Series Forecasting
  470. Unlearnable, or Unmeasured? On the Reliability of Difficulty Labels in RLVR
  471. OverdoseMoE: A Multi-Expert Framework for Opioid Overdose Risk Prediction
  472. AutoDataBench: A Data-centric Testbed for Accelerating Auto Research
  473. GateSPINE: Gated Cross-View Fusion for Lumbar Spine MRI Report Generation
  474. PTNO: Training Neural Operators with Noisy Monte Carlo Estimates for Particle Transport Problems
  475. Replay on Demand: An Emergent Curriculum for Balancing Adaptation and Forgetting in Continued Pretraining
  476. BatSLAM 2.0: Sequence-Verified Sonar Place Recognition in a Robust Pose Graph
  477. The Effect of Tissue Detection on False Positives of Diffusion-Based Artifact Detection in Histopathology
  478. Robust and Learned Online Matching in Growing Trees
  479. Accelerated Algorithm for Sparse Regularized Partial Optimal Transport
  480. Grounding Time-Series Foundation Models in Digital Twin Topology for Predictive Maintenance
  481. Inference Auctions
  482. Community-Driven API and AI Writer Design for Openly Scaling Community Notes
  483. From Tweets to Trades: Analyzing the Influence of Public Mood over Stock Market Performance in Turkiye
  484. LARC: Low-Rank Adaptive Residual Connections for Learning in Frozen Models
  485. Proximal Balancing for Causal Effect Estimation under Unmeasured Confounding
  486. MAGiDiff: Sampling the Photospheric Vector Field from UV/EUV Filtergrams
  487. MGhana-ST: A Low-Resource Speech Translation Dataset for Ghanaian Languages and an Analysis of Multilingual Training Trade-offs
  488. Enhancing Autoregressive Video Generation via Representation Adversarial Distillation
  489. OPTS-TTPO: Enhancing Finite-Sample Policy-Gradient Learning with Tree Search
  490. Efficient Active Auditing of Multi-Group Fairness with Bias Probes
  491. WARP: A Unified Benchmark for Invisible Image Watermarking -- Robustness and Protection Against Attacks
  492. Fenchel Tilting: Weighted Correction for Efficient Finetuning of Generative Models
  493. Who Verifies the Graph? Misspecification Attacks on Causal Action Verification for Language Agents
  494. Amortized Bayesian Inference on Multilevel Models of Arbitrary Structure
  495. Component-Weighted Centroid Search for Exact Incremental BPE
  496. Can We Anticipate Violence? Multimodal Learning from Pre-Incident Behavioral Cues
  497. PINNing the pion: conformal deep learning for Fπ(s) and the (g-2)μ hadronic contribution
  498. Richard: Voice-First Mobile Interaction for Persistent Tasks
  499. Overview of BioASQ 2026: The fourteenth BioASQ Challenge on Large-Scale Biomedical Semantic Indexing and Question Answering
  500. TACTIC: Temporal and Context-Aware LLM Tactical Planning for Roadside LiDAR Attacks
  501. Reconstructing the Dynamic World: A Representation-Centric View of 4D Scene Reconstruction
  502. Coverage Before Control: Route-Instruction Grounding and Steering for Controllable Retrosynthesis
  503. LEAP: Learned Block-wise Evidence Retrieval for Long Audio-Video Perception
  504. Reliability-Aware Checkpoint Selection for Domain Generalization
  505. ConflictGuide: AutoResearch Improves When Competing Behaviors Are Made Visible
  506. RoPE at the End of Its Rope? Theory, Diagnosis, and Mitigation of Long-Context Failures
  507. AdaGEPA: Adaptive Feedback Allocation for Reflective Prompt Optimization
  508. Super-Resolving Unseen Hyperspectral Sensors at Any Scale via Spatial Operators
  509. Cluster Attention Neural Operators for Solving Parametric Partial Differential Equations
  510. TRACE: Trajectory Selection for Parallel Scaling of Search Agents
  511. DoGBench: Can Agents Meet Expert Standards for User-Facing Documentation?
  512. Understanding Parents' Complex Views of AI for Children's Pretend Play
  513. Spatial-Temporal Multi-scale Network for Screen Content Video Quality Enhancement
  514. Shared Weights, Selected Computations: How Looped Transformers Route What Each Loop Does
  515. Markovian Dynamics Enforcer: Feasibility Preserving Correction on Learned Dynamics Manifolds
  516. OPSRD: On-Policy Self-Role Distillation
  517. Grounding with Confidence: Controllable Generative Video Temporal Grounding
  518. PassGPT+: Leveraging Linguistic Priors for Password Modeling
  519. Algorithmic Recourse Under Competition
  520. Completion-Aware Cross-Fidelity Offline-to-Online Reinforcement Learning for Multi-Line Bus Holding
  521. FIGS: Evaluating Multi-Turn Sycophancy Without Penalizing Empathy
  522. Dimension-Free Rank Lifting from Random Hyperplane Arrangements
  523. Predicting Multi-View Rashomon Representation: Can We Learn Where Models Disagree?
  524. BayesNDE: Bayesian Generative Modeling for Neural Density Estimation
  525. DyRAD: Radar Novel View Synthesis for Dynamic Driving Scenes
  526. Dynamic LoRA-Experts and Prototype-Ensemble Matching for Class-Incremental Learning
  527. Learning Steganography Is Easy, Learning Steganographic Reasoning Is Hard
  528. Fast Regularized Policy Mirror Descent with One-Step TD Updates
  529. RainAtlas: A Multi-Continental Dataset for Precipitation Downscaling
  530. P-SRM: Selective Recovery of Rejected Predictions in Visual Tracking
  531. Estimation of the Label-Noise Transition Matrix with Performance Guarantees via Selective Classification
  532. Synthetic Pre-pretraining Survives Scale, but Not as a Grammatical Prior
  533. Should I stay or should I show? Learning to selectively disclose information
  534. Beyond Accuracy: Prefix-Invariant Realizations of Low-Precision Fast Matrix Multiplication
  535. Backward-State Policy Is Part of the Learning Algorithm
  536. A Comprehensive Benchmark of Source-Free Universal Domain Adaptation on Time Series Representations
  537. Stress-Testing LLM Lie Detectors: Role-Play Failures and Spurious Correlations
  538. Finite-Horizon Fisher Memory in Two-Sided Power-Bounded Recurrent Systems
  539. Probabilistic Adversarial Training
  540. TopTimeNet: Topologically-assisted time-series classification model
  541. Pseudo-Label-Triggered Retraining from Forecast Errors for Online Time Series Forecasting
  542. Safety of Latent Communication in Multi-Agent Systems
  543. Seeing as Humans Do: Learning from Motion to Segment Anything Without Supervision
  544. CORD: Learning Reusable Degradation Representations Across Heterogeneous Physical Systems
  545. Riemannian Flow Models with Reinforcement Learning for Molecular Crystal Structure Prediction
  546. How Does Local Landscape Geometry Evolve in Language Model Pre-Training?
  547. MemCodex: Self-Programming Hierarchical Memory for Language Agents
  548. Determining Vertical Displacement of Agricultural Areas Using UAV-Photogrammetry and a Heteroscedastic Deep Learning Model
  549. Validity-Preserving Hierarchical RL for Joint Routing and Switch Placement in EDA
  550. FAST: Flow Any Scene Transformer
  551. The Nixtlaverse: An Open-Source Ecosystem for Forecasting
  552. Stable Transformers for Graph Generation
  553. CNCGEN: A Dataset and Framework for Machining Process Planning and Toolpath Generation from B-rep Models
  554. A library for differentiable signal processing and machine learning on the sphere
  555. Let the Carrier Carry the Attack: Preserving the Subject in Adversarial Image Generation
  556. ArchitectureIQ: On the Measure of Training Intuition
  557. Amortized ratio-estimation importance sampling and localized simulation-based calibration for intractable likelihoods
  558. Drift Inspector: Exploring and Measuring Scientific Drift with Atomic Contribution Claims
  559. A helps B while B hurts A: directed transfer in instruction-tuning mixture
  560. Values as Style: Disentangling Values from Semantics with One-Way Mixing for Low-Damage LLM Steering
  561. Unapologetically Distributed: A Call for Decentralized Document Analysis
  562. MC-PanDA++: Simpler, Stronger, and More Robust Domain-Adaptive Panoptic Segmentation
  563. SE-ADD: Self-Evolving Audio Deepfake Detection with Mistake-Driven Supervision
  564. NodeGround: A Node Classification Benchmark in the Graph Foundation Model Era
  565. The Evolution of Attention in Large Language Models: Mechanisms, Trade-offs, and Emerging Trends
  566. BAM! Bayesian Anything Model: a foundation model for generative computational imaging
  567. Graph Residual Conjugate Diffusion: SNR-Equalized Heat Flow for Graph Signals
  568. FANVIDv2: Evaluating Video Super-Resolution by Face and Licence-Plate Recognition Under Compound Degradation
  569. Beyond Uniform Compression: Budgeted Transmission Allocation for Extreme Federated Learning
  570. SEPAL: Separated Expert Pairs with Answer-Level Fusion for Reliable LLM Collaboration
  571. RiboUnmix: Learning Shared Translational Dynamics from Biased and Noisy Ribo-seq Measurements
  572. Zero-Compute Cross-Lingual Transferability Estimation Using Typological Feature Proxies
  573. Marginal Response Surface Elicitation for Zero-Label Tabular Learning
  574. SAGE: Salient Factor Discovery and Generation with Visual Foundation Representations
  575. Free Everywhere, Exact on Trees: PPO's Dropped Correction Buys Sample Efficiency Under Aggressive Reuse
  576. PEG-Tab: Sampling-Time Record Repair and Release Control for Tabular Synthesis
  577. Certification-Based Differentially Private Learning
  578. Introduction to Computer Vision
  579. Parameterization method of reservoir properties for ensemble-based data assimilation using intermediate latent space of StyleGAN
  580. ExpandDiff: Dynamic Range Expanding Diffusion for Single-Image HDR Reconstruction
  581. Semantic Watermarking for Malicious Image Manipulation Detection
  582. Hybrid Methods for Robust Tabular Data Imputation
  583. Is This Evidence Decision-Critical? Learning to Verify Rule-Governed Decisions
  584. GroundingPI: A Grounding Foundation Model towards Physical Intelligence with Visual Primitives
  585. Text-to-3D Policy: Fine-Grained Language-Behavior Alignment for Unseen Specification Generalization
  586. Convergence of Practical Muon with Finite Newton-Schulz Iterations and Nesterov Momentum
  587. Structural Limits of the Information-Theoretic Uncertainty Decomposition
  588. SPOON: Towards Coherent Compositional 3D Scene Generation from Uncalibrated Multi-view Images
  589. A Generalizable and Explainable Framework for Synthetic Video Detection Using First-Digit Gradient Statistics
  590. Cybersecurity in Edge Computing: A Trust-Aware Federated Hybrid Intrusion Detection Framework
  591. MADGRAV: a multilevel anomaly-detection pipeline for gravitational-wave searches applied to LIGO data
  592. Robust Transfer Learning for Paper ECG Recognition
  593. AVERT-VLN: Abstention-aware Visual Error Recovery and Training for Vision-and-Language Navigation
  594. Compact Language, Complex Model Shifts: How and Where Ambiguity and Underspecification Affect LLMs
  595. Self-Spec Verifiable Code Generation
  596. Invariant Shape Analysis of Surfaces with Spherical Topology
  597. RESUME: Recurrent State Updates from Motion and Residual Signals for Efficient Video Language Modeling
  598. Candidate Retention for Abductive Learning
  599. Divide and Collapse: MAPF-Collapse via Exact Decomposition into Independent Sub-Instances
  600. General Performance Guarantee for Human Torque Estimation-Based Task-Agnostic Assistive Exoskeleton Control
  601. EffGS: Efficient and High-Fidelity Gaussian Splatting
  602. Ghost in the Encoder: Decodable Artist Identity Representations in Lyrics-to-Song Generation
  603. Hyperbolic Prototype Routing for Rehearsal-Free Class-Incremental Learning
  604. Growing an Agent/Prover Interface: Evolutionary Tool Design for Cost-Efficient Theorem Proving in Rocq and Lean
  605. A Reusable Semantic Web Framework for Evidence-Grounded Fundamental Rights Impact Assessments under the EU AI Act
  606. Lens Flare Removal and Reconstruction
  607. Mitigating Representation Gaps in Amortized Bayesian Inference with Auxiliary Supervision
  608. CIDER-FM: Foundation Models for Causal Inference from Diverse Experimental Regimes
  609. Referential Uncertainty in Human--AI Collaboration
  610. Can Domain Generalization be Guaranteed in Small-Sample Learning?
  611. The Geometry of Randomized Smoothing on Feasible Sets
  612. When the Right Answer Is Missing: An Arithmetic-Dependent Rejection Bottleneck in Jev
  613. Towards Robust Time Series Learning via Capacity-Centric Modulation
  614. From Wrecks to Wisdom: Recovering Crash Mechanics from Real-World Multi-View Photos
  615. CAMOS: Coupled Oscillatory State-Space Model for Multimodal Clinical Time-Series
  616. Who Owns That? Evaluating Ownership Intuitions in Large Language Models
  617. About the Influence of Workflow Topology on Task Intensity Prediction through Graph Learning
  618. Beyond the Shadows of Plato's Cave: Evaluating False Memory in Autonomous Agents via Counterfactual Reasoning
  619. DensePed-Lite: Quality-Aware Adaptive Detection for Dense Pedestrians under Occlusion
  620. T-ARC: Topology-Aware Randomized Clustering via Distributionally Robust Stochastic Block Models
  621. Understanding Head Geometry and Dynamics in Federated Regression through a Natural Solution Selection Rule: An Unconstrained Feature Model Analysis
  622. Right-Wing Rock or Just Rock? A Computational Linguistic Analysis of Frei.Wild
  623. Mutual Equilibrium: Multimodal Representation Learning through Reciprocal Feedback
  624. From Speech to Editable Concepts: Probing Emotion Recognition with Concept Bottleneck Models
  625. ResARC: Residual-Aware AutoRegressive Coding for Ultra-Low Bitrate Image Compression
  626. Distributionally robust linear regression through the lens of adversarial training
  627. DuplexAct-Bench: Broadening Full-Duplex Speech Evaluation toward Proactive Interaction across Diverse Behavioral Requirements
  628. Principal Component Regression Dominates all Monotone Spectral Filters for Linear Regression
  629. Towards Trustworthy AI for Glioma Diagnosis: A Task-Aware Evaluation of Uncertainty Quantification
  630. PCB-MC: Missing Component Analysis in Printed Circuit Boards
  631. Awakening of the Buddha: Subspace Learning During Population-Loss Plateaus
  632. Inferring Causal Relations between Two Sequences of Events with Language Models
  633. Towards Optimal Inventory Control under Censored Demand: A Biased Sample-Average Approximation Approach
  634. Can Computation from Earlier Problems Help LLMs Solve New Ones?
  635. Beyond Simulation: Retain-and-Repair Neural Operators for Real-World Adaptation
  636. When, Not How Much: Evaluating Time-Series Foundation Models on Sparse Events
  637. TTLab at Daleel 2026: STAR-Ar, Sequence Tagging for Argument Recognition in Arabic
  638. From Search to Signal: Online Post-Training in Automatic Heuristic Design
  639. InfoAgent: Traceable Generation and Repair of Evidence-Grounded Infographics
  640. EgoTools: Towards Tool-Centric Reasoning in Real-World Egocentric Videos
  641. Beyond the Current Scene: Event-Referential Grasping with Active View Selection
  642. Exploring Heterogeneous Model Merging Approach for Complex Knowledge Transfer
  643. Making Grid Beam Search Less Greedy
  644. A Dynamical Theory of LoRA in Continual Learning
  645. COBICount: Separating Object and Background Responses for Remote Sensing Object Counting Without Training on Target Data
  646. Ready2Blend: From Natural-Language Instructions to Composable Alignment Prompts
  647. LampAttention: Look-Ahead Mixed-Precision FlashAttention for Dedicated Accelerators
  648. Autoresearch in Mixed-Integer Linear and Nonlinear Programming
  649. Working Around the Compute Ceiling: Byte-Exact Memory in Galahad Makes LLM Reading a One-Time Cost LLM Reading a One-Time Cost
  650. Robustifying Asynchronous SGD via Soft Throttling
  651. On the Complexity of Preference-Based Bandits
  652. Who Said What, and Will It Be Remembered? Evaluating Persistent Speaker Attribution Across Meetings
  653. The Golden Path Hypothesis: Reusable Schedules in Diffusion Caching
  654. Understanding as No-Arbitrage: Bounded Dutch Books as a Definition and Training Objective for Language Models
  655. ElectrolyteFM: Unifying Electrolyte Property Prediction through Cross-Property Knowledge Learning
  656. A Unified Dual Method for Matching Problems
  657. How Many Samples Are Enough for Learning Across Domains?
  658. TexTailor: Texture-Preserving Video Virtual Try-On via Adaptive Garment Conditioning
  659. NarrativeSteward: Coordinating Delegation, Guidance, and Verification in Agent-Assisted Interactive Narrative Authoring
  660. Discrete Score Matching Enables Causal Discovery from Count Data
  661. HiWE: Hierarchical World Knowledge Model with Visual Keypoint Enhancement for Zero-Shot 3D Path Planning
  662. Rethinking Generative Image Compression at Extremely Low Bitrates
  663. Generalized Geometry Block Proximal Linearized Method for Multiblock Nonconvex and Nonsmooth Optimization
  664. BMASH: Ball-Motion-Aware Soccer Header Spotting
  665. Semantic-Aware Joint Source-Channel Optimization for Encoder-Agnostic Digital Video Communication
  666. Physics-Informed Method of Group Data Handling: Adaptive Construction of Functional Representations with an Application to the Navier-Stokes Equations
  667. EngramBench: A Capability-Grounded Benchmark for Skill-Evolution Harnesses
  668. A Tilted Bowl Is Not a Slippery Slope: Compressing Looped Models
  669. MegaAvatar: Controllable Talking Avatar Generation
  670. RW-Flow: One-Step Generation on Compact Manifolds via Wasserstein Gradient Flows
  671. A Time-Aware Bag-of-Receptive-Fields for Interpretable Irregular Time Series Classification
  672. PLRS-IC: A Dual-Calibration Framework for Chest X-Ray Vision-Language Alignment
  673. Universal Cross-Prompt Adversarial Attacks on Promptable Concept Segmentation
  674. Jacobian Rank Collapse in Decision-Focused Learning
  675. Effective Does Not Mean Useful: Conditional Functional Substitutability for Redundancy and Scaling in Transformers
  676. From Benchmarks to Production: Transferring Time Series Anomaly Detection Methods for Electricity Production Monitoring
  677. Client and Training Data Selection for Computationally Efficient Synchronized Federated Learning
  678. Trust the Critic More
  679. A differentiability framework for zigzag persistent homology via linear interpolation
  680. 4MT-VLM: How Coarse Is a VLMs Cognitive Map?
  681. What Streaming Anomaly Detection Finds (and Misses) in Industrial Time Series
  682. RAIM: Robust Aggregation of Inexpensive Models for Hallucination Detection
  683. Fyan: A Human--AI Harness with Semantic Auditing for Document-Level Formalization
  684. Argument Structure Prediction in Online Conversations: A Comparative Study of Modeling Paradigms and Task Architectures
  685. QATFactory: A Versatile, Deployment-Aligned Framework for Quantization-aware Training and Distillation of LLMs
  686. In a Streaming World, Should You Stand Still? A Comprehensive Benchmark of Anomaly Detection in Streams
  687. Minimax Additive Regression under Unknown Dependent Designs
  688. Uruqi: Learning Spatial Cognition from Visual Experience
  689. Importance-Aware Feature Sparsification for Wireless Split Learning
  690. Dynamics to decision: A mathematical theory of Lyapunov spectra and decision boundaries in deep classifiers
  691. ViLegalExpert: A Large-Scale Benchmark for Vietnamese Legal Retrieval and Question Answering from Real-World Consultations
  692. Attention Function as an Intrinsic Inductive Bias: How Models' Behavior Diverges in Novel Contexts
  693. Fiber-Resolved Microstructure Quantification from Multi-Shell Diffusion MRI using Detection Transformers
  694. Asymptotic Properties of Support Vector Machines in High-Dimension, Low-Sample-Size Settings under a Spiked Model
  695. A Width-Matched Comparison of Hybrid Quantum-Classical Self-Supervised Learning for Fingerprint Recognition
  696. Deep Learning-Based Tri-Hybrid Multi-User MIMO Precoding: The Blessing of EM-Reconfigurable Antennas
  697. Beyond the Remembered World: Predictive 4D Belief for Persistent Navigation in Evolving Worlds
  698. QuanVI: Score-based Variational Inference via Quantum Maximally Mixed States
  699. Training-Free Affinity Fusion of Neural and Embedding-Based Speaker Diarization
  700. TripleFlow: Training-Free Video Object Removal by Bridging Residual Editing and Native Generation
  701. Do Self-Evolving Skills Generalize to Held-Out Tasks?
  702. MindWorldBench: Evaluating Mental-State-to-Behavior Reasoning in Image-to-Video Generation
  703. Sharp Stationary Gaussian Approximation for Constant-Stepsize SGD
  704. RefCon: Iterative Refinement and Contrastive Memory Extraction for Context-Evolving Agent
  705. BELIEFRAG: Making Adaptive RAG State-Aware under Evolving Evidence
  706. Uncertainty-Aware Consistency Distillation for Few-Step Video Generation
  707. Perceptual Color Difference Modeling Using Machine Learning and Human Similarity Judgments
  708. GeoGAT: Bidirectional Temporal Sampling Meets Hierarchical Graph Attention for Global Video Geo-localization
  709. How to Reduce Localization Ambiguity? Geometry-Semantic Constrained BEV Representation Learning for Satellite-Ground Localization
  710. Prequential E-Values for Selected-GP Near-Optimality Certificates
  711. GRC-Pose: Generation-Reconstruction Correspondence for Prior-Free 6D Object Pose Tracking
  712. Beyond Local Linearity: Scale-Resolved Geometry of Learned Image Encoders
  713. The Row Normalization Puzzle in Muon
  714. Bongard: Training Machine Intuition
  715. T-Router: Learning Thalamic Routing for Reasoning with Parameter-Efficient Reinforcement Learning
  716. Reserve-Aware Contrast Certificates for Conservative Bandits with Uncertain Baselines
  717. False Frontiers: Diagnosing and Mitigating Co-Cheating in Self-Evolving Search Agents
  718. A Generalisation Signal Need Not Be a Model-Selection Signal
  719. Steepest Guidance: A Practical and Principled Approach to Inference-Time Alignment of Flow and Diffusion-based Models
  720. A strategic roadmap for an atomistic machine-learning ecosystem
  721. UGOD: Uncertainty-Guided Opacity and Dropout for Sparse-View 3D Gaussian Splatting
  722. SCIC: Scope- and Codebook-Aware Instruction Conditioning for Speaker-Adapted Expressive TTS
  723. Trustworthy Runtime Error Healing in Real-World Repositories: A Benchmark and Guardrail
  724. MRI Super-Resolution with RCDM/WaveMix and Task-Aware Segmentation
  725. Shared Phase and Retention Control for Efficient Adaptive Spectral Recurrence
  726. Association profile conditioning in a set-temporal transformer for cross-session intracortical motor decoding
  727. Parameter symmetries determine representational geometry in overparameterized nonlinear networks
  728. HO-FL: Hybrid-Order Federated Learning for Heterogeneous Edge Devices
  729. Beyond Text: LLM-Based Dimensional Emotion Evaluation in Multimodal Dialogue
  730. Argus: A Real-EKS Study of When Predicting Spot Interruptions Beats Simple Checkpointing
  731. A 3GPP-Compliant Benchmark Dataset for RIS-Aided Beyond 5G Networks
  732. The Missing Coefficients: Bayesian Pairwise Merging for Model Personalization
  733. TSMD: Temporal-Stream Modality Dropout for Robust Video Highlight Detection
  734. Structure vs. Chain-of-Thought: Evaluating LLM Criteria Extraction for Depression Severity
  735. Structure-aware Reinforcement Learning for Protein Directed Evolution
  736. Hard-Gate Candidacy in a Deployed Validator Suite
  737. Cycle-Aware Autoencoder with Cross-SignalConsistency for Railway Door Anomaly Detection
  738. Switching Linear Attention
  739. A Missing Piece for Trustworthy AI Reviewers: From Benchmarking Rhetorical Robustness to SciCore Review
  740. A Rank Graduation metric for Algorithmic fairness
  741. Minimax rates for learning spectral Barron functions by deep ReLU neural networks
  742. Synchronous Multi-view Neural Diffusion
  743. Make Code as Policy Great Again: Frontier Agents Write, Call, and Evolve Robot Tools
  744. Evidence First, Arithmetic Second: A System Report and Failure Analysis for DocSem
  745. An Uncertainty-Guided Digital Twin Framework for Online Adaptive Proton Therapy in Head and Neck Cancer: A Feasibility Study
  746. C-STRIDE: An Observation-Driven AI Digital Twin for Predicting Basin-Wide Flood Fields from Sparse Stream-Gauge Histories
  747. When Integral Meets Decomposition: A Signal-Level Self-Supervised Feature Decompose Paradigm for Multi-Modal Image Fusion
  748. The Invisible Language Tax: Token Premiums of French and Regional Languages in 2026 LLM Tokenizers, and a French-Optimized Prototype
  749. Not all solutions are created equal: An analytical dissociation of functional and representational similarity in deep linear neural networks
  750. Settle: Learning When to Stop Reasoning
  751. Anchoring Adversarial Trajectories to Data Manifolds: A Bilevel Transfer Optimization Framework
  752. MeshOctave: Vertex Split-and-Rewire Cascades for Native Mesh Generation
  753. Approval Laundering: Systematizing Approval--Execution Binding Failures in AI Coding-Agent Harnesses
  754. SCORE-LM: State-Space Radar Representations with Language Models for Fault Diagnosis
  755. Scale-Split Neural Operator for Memory- and Data-Efficient 3D Turbulence Prediction
  756. Fairness Beyond a Single Run: Training-Seed Variability in Speech LLM Adaptation
  757. Beyond Spatial-Domain Supervision: A Relation Constrained Space for Multi-Modal Image Fusion
  758. Routing Probes Can Improve Without New Information: An Exact-Null Audit of Uncertainty Beyond Model Outputs
  759. APTInvestBench: Evaluating Autonomous APT Investigation under Varying Telemetry
  760. Signal-Routed Temperature Scaling: Low-Capacity Risk-Conditioned Calibration for Small Validation Budgets
  761. PrivCert: Certifying Statement Support under Differential Privacy
  762. EFormer: Temporally Aligned Local Correction for Continuous sEMG-Based Hand Pose Tracking
  763. On the Relaxation of Conditional Independence Assumption for Image Segmentation
  764. PrecipJEPA: JEPA-Regularized Future-State Prediction with Motion-Source Rendering for Precipitation Nowcasting
  765. Prototype-guided Bilateral Alignment Multimodal Federated Learning
  766. GraphForge: Training Working Agents with Graph-Anchored Workspace Synthesis
  767. Learning Where to Steer: Noise-Space Geometry for Efficient Offline Multi-Objective Optimization with Generative Models
  768. How Much Can Reliability Drift Under a Fixed Confidence Distribution?
  769. Warm-starting PDE solvers with any-dimensional machine learning
  770. Risk-Aware Adaptive Evaluation: Finding High-Impact Failures Under Limited Budgets
  771. FLOW: Feature-Level Optimal Warping for Generalized Remote Physiological Measurement
  772. Generalized Residual Closure: General Learning Dynamics for Stability-Plasticity Compatibility
  773. CellMSA: Context Modeling for Single-Cell Representation Learning
  774. Characterizing Questioning Patterns and Student Engagement Through Contextual Analysis of Real-Time Classroom Interactions
  775. Amortized Data Borrowing with Exchangeability-Aware Neural Posterior Estimation
  776. Certified Approximation for Interpretable Representer Landmarks
  777. MEMO: Multi-Level Entity-Aware Memory for Streaming Video Understanding
  778. FFASR: Benchmarking Far-Field Automatic Speech Recognition using High-Fidelity Simulated RIRs
  779. Learning Continuous Neural Representation of Stochastic Hybrid Systems
  780. VERA: Verifiable Feasibility Representations with Counterfactual Credit for Constrained Multi-Agent Control
  781. VOSSA: Voiceprint Optimization for Streaming Speech Architectures
  782. Right Answers, Costly Models: The Efficiency Gap in LLM-based Optimization Modeling
  783. Near-Minimax-Optimal Rates for Asynchronous TD Learning with Markovian Data
  784. Audio Token Attention Is Predictable Before the Language Model Runs
  785. On Parameters of Nonlinear Scalar Dynamics from Video: Invariants, Calibration, and Identifiability
  786. Optimal Allocation and Volume under Surface
  787. Reasoning Externalization for Faithful Large Language Model Narratives of Stock Return Predictions
  788. AdaOcc: Adaptive 3D Occupancy Prediction for Embodied Tasks
  789. GeoNest: Learning to Select Failure-Aware Neighborhoods for the Irregular Knapsack Problem in a Circular Container
  790. TRACE: Target-Aware Retrieval, Attributed Evidence, and Contract-Constrained Extraction for LitTraceQA
  791. Decoupling Spherical Reasoning from Dense Prediction for 360 Depth Estimation
  792. Where MLLMs Fail and Why: Causal Task Decomposition for Capability Failure Diagnosis
  793. Learn-Then-Differentiate Gradient Estimation
  794. scTrilemma: Balancing Identity, Invariance, and Fidelity in Single-Cell Representation Learning
  795. FrameMorrow: Future-guided Frame Selection with Prospective Tokens for Long-Horizon Video Generation
  796. Average-and Last-Iterate Lower Bounds for Optimistic Matrix Mirror-Prox in Quantum Zero-Sum Games
  797. Optimal VC Dimension of Contrastive Learning with Margin
  798. ReSCENE: Server-Side Replay for Structural Mitigation of Catastrophic Forgetting in Federated Continual Learning
  799. Scaling Parameter and Context in Attention: Native Sparse Attention from Mixture-of-Head
  800. Forging LLM Authorship Fingerprints with Targeted Rewriting
  801. More Choices, Fewer Decisions: Ordinal-Scale Bias in JEV-like Direct-Decision Models
  802. BARRAC: Adaptation of an English Aspect-based Sentiment Analysis Approach for Classification Tasks in Arabic Dialects
  803. Whose Voice Survives the Summary? A Voice-Retention Audit of LLM Employee Listening
  804. You're Hired: Strategic Model Selection for LLM Collaboration
  805. Learning Chaos Without Seeing Chaos: Extrapolation of Global Dynamics in Autoregressive Transformers
  806. Can Terminal Agents Trust Their Own Verification? Diagnosing and Improving Self-Verification
  807. DCM-SAM: Defect-Conditioned Mixture of LoRA Experts for NPU-Deployed AM Defect Segmentation
  808. Uncovering Uncontrolled Repetition through Residual Stream Dynamics
  809. Overlap, Unique and Conflict: Can LLMs Extract What They Can Recognize?
  810. Evaluating Persistent Calibration under Evolving Model Knowledge
  811. LEARN-TS: LLM-Enhanced Alignment and Reconstruction with Normality Guidance for Multivariate Time-Series Anomaly Detection
  812. Positive Ratings, Hidden Concerns: Employee Voice Disclosure in AI-Mediated Organizational Listening
  813. Same Loss, Different Gradients
  814. How Accurate Is Accurate Enough?
  815. Does a Shared Temperature Imply a Shared Angular Scale in Probabilistic Contrastive Learning?
  816. RAST: Resolution-Aware Privileged Structure Transfer for Low-Resolution Audio Activity Recognition
  817. Learning Under Forgetting: Statistical Support-Selective Retention in Stochastic Training Dynamics
  818. dattri-LLM: A Unified and Efficient Library for Training Data Attribution at LLM Scale
  819. PathAnchor: Path-Structured Evidence for Scientific Agents
  820. Lasting Effects of Abstract Pretraining Beyond Perplexity
  821. Adaptive-GEPA: Make Your Harness Fit Heterogeneous Requests
  822. Event-Driven Refresh and Recurrence Memory to Reduce Stale Grounding in Referring Video Object Segmentation
  823. Agentic Relative Camera Pose Estimation via Learned Ranking and Verification
  824. Where the Evidence Lives: Auditing AI Companions' Self-Descriptions
  825. Hear the World in Stereo: Learning Dynamic Spatial Correspondence for Immersive Joint Video-Audio Generation
  826. Consensus-Aware Multi-Source Fusion for Reference-Guided Camouflaged Object Detection
  827. Matisse: Evidence-Space Reasoning for Active 3D Reconstruction
  828. Molecular Property Prediction under Structural Shift with Tabular Foundation Models
  829. MetaSteer: Context-Conditioned, nonlinear Steering via Attention-Projection Adaptation
  830. Hard-Region Supervision: #1 on the Waymo Open Dataset 2D Video Panoptic Segmentation Leaderboard
  831. SCALE: Synthetic Calibration via Agreement Labeling in Embedding Space
  832. Always-On Experimentation
  833. No Corners Cut: State-Grounded Transitions for Mid-Stream Prompt Switches in Video Generation
  834. GATE-ST: Gene-Aware Text-image Encoder for Spatial Transcriptomics
  835. EPIC: Epipolar-Consistent 360deg Immersive Stereo Video Generation
  836. Concept-Grounded Attention: A Controlled Evaluation of Graph-Injected Attention, Temporal Versioning, and Epistemic Status
  837. Unveiling the Value of Motion for Cinematic Camera Trajectories