mlsanityData to 14 Oct 2026

Papers

New on arXiv

Fri 2 Oct 2026

601 papers

  1. Feature Elicitation for SAE Steering of Deceptive Behaviour in Small LLMsMechanistic interpretability → Probing deception, bias and beliefs
  2. Reinforcement Learning from Real-Robot Interaction without High-Quality State EstimatesDeep reinforcement learning methods → RL for robot locomotion and manipulation
  3. A Geometric Perspective on Reward Function Updates in Inverse Reinforcement LearningOffline RL and imitation learning from data → Inverse reinforcement learning +1
  4. RAIN: Robotic Task Generalization with Region-Aware Interaction NetworksMultimodal LLMs / vision-language models → Embodied and 3D vision-language models
  5. RadOnc-Agent: An LLM-Orchestrated Framework for AI Workflows Across the Radiotherapy Care PathwayLLM agents and tool use → Medical and scientific domain agents
  6. Does a Learned Corrector Beat a Simple Retreat? Evidence from a Frozen VLAVision-language-action models → Failure recovery and robustness
  7. Learning from Unreliable Trajectories: Adversarially-Robust Federated Q-LearningDeep reinforcement learning methods → Adversarial robustness and RL in finance
  8. FluidPD: In-Place Elasticity for SLO-Aware Prefill-Decode Disaggregated LLM ServingEfficient LLM inference and serving → Serving systems and request scheduling
  9. GAMEGO: Training Game-Dev Agents with Synthetic Trajectories Anchored in Real-World AssetsLLM agents and tool use → Software engineering coding agents
  10. Trajectory-Derived Confidence for Reliable, Resource-Aware Clinical Text-to-SQL AgentsLLM agents and tool use → Medical and scientific domain agents
  11. Kepler4D: Controllable Future Video Generation via 4D Scene State EvolutionWorld models → Video-generation world models and simulators
  12. ExpertMuon-Compass: Alignment-Guided Step Sizes for Mixture-of-Experts TrainingMixture-of-experts → MoE large language models
  13. From Sight to Foresight: Predictive Spatial Reasoning in Vision-Language ModelsLLM reasoning → Spatial and embodied reasoning +1
  14. Dynamic Harness Search: Building Multi-Agent Systems Per-Query via PredictionLLM agents and tool use → Software engineering coding agents
  15. Auditing Pairwise Equivalence Judgments: Self-Critique Effects and Diversity Measurement in Multi-Agent Hypothesis GenerationLLM agents and tool use → Social simulation and multi-agent debate
  16. PB-GRPO: Learning Socially Adaptive LLM Agents from Persona-Driven Simulation with Preference-Batched GRPOLLM reasoning → Reinforcement learning for reasoning +1
  17. InvestigationWorlds: An Agentic Environment for Legal InvestigationLLM agents and tool use → Web, search, and research agents
  18. Copying Before Suppression: What Drives a Below-Chance Dip During Language Model Training?Mechanistic interpretability → Knowledge and representations inside LLMs
  19. CUAWright: A Minimal Unified Interface for Digital AgentsLLM agents and tool use → GUI and mobile agents
  20. The Independence Prior of SAEs Fragments Visual ConceptsMechanistic interpretability → Sparse autoencoder training and evaluation
  21. SEER: Self-Evolving Event Reasoning and Retrieval for Time Series ForecastingLLM reasoning → Time series reasoning
  22. Representation-Aligned Auxiliary Supervision for Language Model AdaptationLLM reasoning → Reinforcement learning for reasoning
  23. Pareto-Dominant Clarification: Post-Training Coding LLMs via PPO-Lagrangian Budget ConstraintsLLM agents and tool use → General LLM agent frameworks and benchmarks
  24. Towards Safer Autonomous Driving in an Open World: A Dual-Process ApproachEnd-to-end autonomous driving → End-to-end system architecture
  25. DUET: Co-Evolving Solver and Grader AgentsLLM agents and tool use → Software engineering coding agents
  26. Self-Propagating Misalignment in LLM Agents, and Why Auditing or Disabling Memory Is Not EnoughLLM agents and tool use → Agent threats, guardrails and governance
  27. IdeaScientist: Orchestrating Agents for Grounded Scientific IdeationLLM agents and tool use → Web, search, and research agents
  28. Reinforcement Learning with Comparative Evidence for Social IntelligenceDeep reinforcement learning methods → RL training of language and agent models
  29. Discrete Diffusion for Large Graph Generation via Structural Candidate RestrictionDiffusion models → Diffusion foundations and sampling
  30. Reward-DAgger: Robot-Gated Interactive Imitation Learning with General-Purpose Progress-Based Reward ModelsLLM alignment / RLHF / preference optimisation → Reward hacking and misspecification +1
  31. Agent Policy-Value Audit: Separating Transition Composition from Event Selection in Financial LLM AgentsLLM agents and tool use → Financial trading and forecasting agents
  32. Protecting Sensitive Data in Image Synthesis via PAC-Private Adaptation for Diffusion ModelsDiffusion models → Robustness, privacy and adversarial aspects
  33. ClasSAE: Class-Aligned Sparse Autoencoders via Differentiable Feature-Class AffinityMechanistic interpretability → Sparse autoencoder training and evaluation
  34. Slaying the Hydra: Interaction-Aware Circuit Discovery in Language ModelsMechanistic interpretability → Circuits and mechanistic analysis
  35. PhysMamba: Selective State Space Models as Learned Articulated Body SimulatorsState-space sequence models (S4, Mamba) → Video, motion and generative Mamba models
  36. SUAVE: Unified Video-Action Models via Masked DiffusionWorld models → Embodied and robot world models +2
  37. SkillScriptBench: Benchmarking Self-Evolution of Executable Agent Skill Packages Beyond MarkdownLLM agents and tool use → Agent skill learning and evolution
  38. Teaching Agents to Code ReliablyLLM agents and tool use → Software engineering coding agents
  39. FADE: Frame-Aware Diffusion-Transformer-based Multi-Concept Erasure for Video UnlearningDiffusion models → Video generation +1
  40. LatentQuant: Preserving the Policy-Facing Latent Contract under NVFP4 VAE QuantizationWorld models → Latent predictive world models
  41. Retrieval-Augmented Large Language Model Decision-Making for Autonomous Driving Guided by Chinese Philosophical WisdomRetrieval-augmented generation → Agentic RAG, agents and memory +1
  42. Dynamic Time Step Prediction in Inverse Heat Dissipation for Blur-Like Image Restoration TasksDiffusion models → Inverse problems and image restoration
  43. MLLMs Fail to Refuse when Using Tools AgenticallyLLM reasoning → Reasoning for agents, planning and code +2
  44. Barrier-Shaped Recurrent Reinforcement Learning for Autonomous Landing on a Heaving Ship DeckDeep reinforcement learning methods → RL for robot navigation and UAV control
  45. When Evidence Changes: Evaluating Memory Repair and Re-reading in Language-Model AgentsLLM agents and tool use → Agent memory systems
  46. Proxy Confidence: Auditing Black-Box LLM Agents with a Surrogate's Log-ProbabilitiesLLM agents and tool use → Tool learning and function calling
  47. Reinforcement Learning on the Discrete Composition Channel of a Crystal Generator: Validated Gains and Reward HackingDiffusion models → Flow matching, time series and science +1
  48. Beware EviLLM: Enabling Vulnerability Injection via Large Language ModelsLLM safety and security → Prompt injection and agent security
  49. Probability flow ODEs in score-based and reflected diffusion modelsDiffusion models → Diffusion foundations and sampling
  50. KVE-KD: Key Visual Evidence-Guided Knowledge Distillation for Vision-Language ModelsMultimodal LLMs / vision-language models → VLM benchmarks and visual reasoning +1
  51. SYNLAT: Syntax-Aligned Text-Latent Compression for Chain-of-Thought ReasoningLLM reasoning → Test-time compute and efficient thinking
  52. Memory-State Critic for Asymmetric Actor-Critic with Application to Vision-Based Pursuit-EvasionDeep reinforcement learning methods → RL theory and sample complexity
  53. From Requirements to Attack Trees: Grounded LLM Agents for Design-Time Security ReviewLLM agents and tool use → Agent threats, guardrails and governance
  54. What Should World Models Forget? Stratified Retention for Continual AdaptationWorld models → World models for language agents
  55. Decoding the Functional Roles of Register and High-Norm Patch Tokens in Vision TransformersMechanistic interpretability → Interpreting vision and multimodal models
  56. Pivot-SD: Efficient Self-Distillation for Masked Diffusion Language ModelsDiffusion models → Discrete and text diffusion
  57. IDRF: Inverse-Distilled Reward Fine-tuning of Masked Discrete Diffusion ModelsDiffusion models → Diffusion foundations and sampling
  58. Credit Where It Matters: Dependency-Aware Policy Optimization for Terminal AgentsDeep reinforcement learning methods → RL training of language and agent models
  59. World Embedding BenchmarkWorld models → Video-generation world models and simulators
  60. NeutronGym: Physics-Graded Neutron Instrument Design for LLM AgentsLLM agents and tool use → Educational agents and math problem solving
  61. CORNAV: Construction-Aware Reasoning for Robot Navigation on Active WorksitesLLM reasoning → Spatial and embodied reasoning
  62. UniIntervene++: An Adaptive Intervention Agent for Efficient Real-World Reinforcement LearningDeep reinforcement learning methods → Deep RL algorithms and representations
  63. DEPICT: Scoring Text-to-Image Alignment by Answer AgreementLLM alignment / RLHF / preference optimisation → Diffusion and multimodal model alignment
  64. Mastering Atari 2600 Games with Discovered OptionsDeep reinforcement learning methods → RL for games and combinatorial optimization
  65. When a Correct Reward Is Not Enough: Diagnosing and Guiding PPO in an Analytically Solved Broker-Trader GameDeep reinforcement learning methods → Adversarial robustness and RL in finance
  66. HazardWeaver: Scientific Route Selection for Hazard Analysis AgentsLLM agents and tool use → Web, search, and research agents
  67. AVL-JEPA: Preventing Causal Dynamics Information Collapse In Joint Embedding Predictive Architecture World ModelsWorld models → Latent predictive world models
  68. Threat-Preserving Representation Sensitivity in Agent-Security BenchmarksLLM agents and tool use → Agent threats, guardrails and governance
  69. Rethinking What to Cache in Few-Step Diffusion Transformers: Solver-Aware Target SelectionDiffusion models → Diffusion foundations and sampling
  70. Writerslogic at PAN 2026: Process over Content for Robust Detection under Domain ShiftLLM reasoning → Reasoning benchmarks and evaluation
  71. Divergence controls entropy in distillationKnowledge distillation → On-policy and self-distillation for LLMs
  72. Reasoning Models Are Accurate but Unsound on IdentificationLLM reasoning → Causal reasoning in LLMs
  73. XGenAct: Geometry-Enhanced World Action Models through Cross-Task GenerationWorld models → Embodied and robot world models
  74. Learning from Repaired Reasoning: Root-Cause-Guided On-Policy DistillationLLM reasoning → Prompting and chain-of-thought reasoning +1
  75. Efficient Reasoning Training Does Not Always Harm CoT Faithfulness and MonitorabilityLLM reasoning → Prompting and chain-of-thought reasoning
  76. Getting Your Guidance Weights Right in diffusion and flow-matching posterior samplingDiffusion models → Flow matching, time series and science
  77. Certified Mechanistic Edits: Behavioral Guarantees for Skill Removal and PreservationLLM safety and security → Machine unlearning +1
  78. Detect and Suppress: A Mechanistic Defense against Adversarial Patches in VLA ModelsVision-language-action models → Safety and adversarial attacks +1
  79. AREX: Affine-Residual Exponential Integrator for Few-Step Sampling in Flow MatchingDiffusion models → Flow matching, time series and science
  80. MobiAgent: Dual-Loop Recursive Policy Self-Improvement for Long-Horizon Mobile ManipulationSynthetic data and self-improvement for LLMs → Self-evolving LLM agents
  81. Single or Multiple Policies for Phase-Structured Reinforcement Learning?Deep reinforcement learning methods → RL theory and sample complexity
  82. A Vision-Language Model (VLM)-based Pipeline for End-to-End Procedural Modeling of Field-Grown Maize from Point CloudsMultimodal LLMs / vision-language models → Agricultural vision-language models
  83. A Near-Zero Monitor Readout Is Not Evidence of Behavioral ControlLLM alignment / RLHF / preference optimisation → Reward hacking and misspecification
  84. Passing the Test You Trained On: Re-evaluating Prompt-Injection Detectors for LLM AgentsLLM agents and tool use → Agent threats, guardrails and governance
  85. Corrupted but Correct: Why Vision-Language Models Lie to Themselves InternallyMultimodal LLMs / vision-language models → MLLM jailbreaks and adversarial robustness +1
  86. OuroReward: Sequential Reward Scheduling for Reinforcement Learning in Text-to-3D GenerationDeep reinforcement learning methods → RL training of language and agent models
  87. CLIMB: Confidence-Guided Complementary Evidence for Multimodal Retrieval-Augmented GenerationLLM reasoning → Multimodal reasoning +2
  88. Iterating Consistency Models: Stability, Error Bounds and Noise SchedulesDiffusion models → Diffusion foundations and sampling
  89. Beyond Entropy: Self-Diagnostic Multi-Role Token Optimization for Video ReasoningLLM reasoning → Reinforcement learning for reasoning
  90. A Unified Framework for Bayesian Data Assimilation with Generative Models and Observation InterpolantsDiffusion models → Flow matching, time series and science
  91. Bidirectional Voronoi-biased Exploration Curriculum for Reinforcement LearningDeep reinforcement learning methods → Deep RL algorithms and representations
  92. Native Action-Prior Learning from Videos for World Action ModelsWorld models → Embodied and robot world models +1
  93. DriftTTS: Few-Step Text-to-Speech Without Distillation via Distribution-Matching DriftDiffusion models → Audio, speech and music +1
  94. From Patching to Pruning Visual Computation in Vision Language ModelsMultimodal LLMs / vision-language models → Visual token pruning and efficiency +1
  95. EVEWorld: Physical Evolution Supervision for Embodied World ModelsWorld models → World models for navigation
  96. Mixture-of-Experts for Cryptocurrency Order Execution: Training Stability, Tail Risk, and Failure ModesMixture-of-experts → MoE theory and gating functions
  97. Follow the Winners: Conservative Policy Improvement with the Cross-Entropy Method for Critic-Free RFTLLM reasoning → Reinforcement learning for reasoning +1
  98. ReFract: Benchmarking Perspective Awareness in Language Model Agents with Text World ModelsWorld models → World models for language agents +1
  99. Equivariant Visual-Tactile Diffusion Policy for Contact-Rich ManipulationDiffusion models → Diffusion and flow policies for robotics +1
  100. SyntaxBench: A Statistical Diagnostic Framework for Character-Level Reasoning in Large Language ModelsLLM reasoning → Reasoning benchmarks and evaluation
All 601
  1. Preserving Mathematical Reasoning in Compressed Diffusion Language Models via Trajectory-Aware Low-Rank ApproximationLLM reasoning → Diffusion reasoning and test-time scaling +1
  2. Refinement Buys Intelligibility, Search Buys Identity: What Test-Time Compute Buys in Masked-Diffusion TTSLLM reasoning → Diffusion reasoning and test-time scaling +1
  3. Lightweight, Rubric-Guided Trajectory Evaluation for Production AI AgentsLLM agents and tool use → Software engineering coding agents
  4. DAWIS: Data Assimilation with Windowed Inverse Sampling via Multitask InterpolantsDiffusion models → Flow matching, time series and science
  5. Architecture-Dependent Fusion Pathways in MLLMsMultimodal LLMs / vision-language models → MLLM training and visual instruction tuning
  6. SPEAR: A Spectral-Disentangled MoE Neural Operator with Knowledge-Guided Expert Aggregation for Large-Scale PDE PretrainingMixture-of-experts → MoE theory and gating functions
  7. Consecutive Posterior Fusion for Diffusive Recovery of Unobservable Image StructuresDiffusion models → Inverse problems and image restoration
  8. D2K-Bench: Can LLM Agents Turn Expert Designs into Efficient GPU Kernels?LLM agents and tool use → Software engineering coding agents
  9. Uncertainty as a Proxy for Semantic Correctness in Diffusion-Based Medical Image SynthesisDiffusion models → Medical imaging and segmentation
  10. AdaStep: Adaptive Step Credit Weighting for Agentic Reinforcement LearningLLM agents and tool use → Reinforcement learning for LLM agents
  11. VDOT++: Unified Few-Step Video Generation via Unbalanced Optimal Transport DistillationDiffusion models → Video generation
  12. Toward SLM-based agentic task-tool intent matchingLLM agents and tool use → Tool learning and function calling
  13. Contextual Flow Matching: Adaptive Step Selection in Flow Models for Efficient Visual GenerationDiffusion models → Video generation
  14. KV2: A Self-Refining KV CacheEfficient LLM inference and serving → KV cache optimization
  15. Source Preference in the Wild: How LLM Agents Favor Items by Source, and How to Reduce ItLLM alignment / RLHF / preference optimisation → Human values and AI feedback alignment
  16. Gains and Collapse in On-Policy Distillation:A Reinforcement Learning PerspectiveLLM alignment / RLHF / preference optimisation → Reward hacking and misspecification +2
  17. Hindsight-Guided Rationale Distillation for Rare Disease DiagnosisLLM reasoning → Clinical and biomedical applications
  18. Sample complexity of variance-reduced policy gradient: weaker assumptions and lower boundsDeep reinforcement learning methods → RL theory and sample complexity
  19. Predicting Steering Vectors and Adapter Weights for Few-Shot Author-Style TransferMechanistic interpretability → Activation steering methods
  20. Does Physics Live in the Activations? Localizing Physical Quantities in Video Diffusion ModelsWorld models → Video-generation world models and simulators +1
  21. EvoRiskBench: An Evolving Benchmark for Runtime Security Risks in Workspace AgentsLLM agents and tool use → Agent threats, guardrails and governance
  22. Behavior Pack Optimization for Video MLLM Post-TrainingMultimodal LLMs / vision-language models → Video understanding with MLLMs
  23. Keeping JEPA World Models Plannable When Little of the Frame MovesWorld models → Latent predictive world models
  24. Investigating the Role of Reasoning-Language Alignment in Monolingual Retrieval-Augmented GenerationLLM reasoning → Reasoning benchmarks and evaluation +1
  25. Page-EntroKV: Hardware-Aligned, Entropy-Weighted KV-Cache Eviction under Grouped-Query AttentionEfficient LLM inference and serving → KV cache optimization
  26. Safe Streaming Flow Planning by Aligning Sampling Dynamics with Execution DynamicsDiffusion models → Diffusion and flow policies for robotics
  27. The Fragility of Trigger-Tag Mechanisms for Misuse Detection in Open-Weight LLMsLLM safety and security → Backdoor and data poisoning attacks
  28. Foresight: planning future perception in streaming VLMs without retrainingMultimodal LLMs / vision-language models → Embodied and 3D vision-language models
  29. In-Distribution Forcing for Long Video Generation at Test TimeDiffusion models → Video generation
  30. How to Find and Reuse Policies for Continuous Adaptation in Lifelong Reinforcement LearningDeep reinforcement learning methods → Deep RL algorithms and representations
  31. Ask, Relax, or Act? Evaluating Actionable Indeterminacy in LLM Preference ReasoningLLM reasoning → Reasoning for agents, planning and code +1
  32. Securing Computer-Use Agents Against Branch Steering AttacksLLM agents and tool use → Agent threats, guardrails and governance +1
  33. Learn Feasibility Once, Optimize All Objectives: Derivative-Free Diffusion Models for Chance-Constrained ProgrammingDiffusion models → Flow matching, time series and science
  34. Where to Look Is Not How to Fix: Pre-Denoising Diagnostics and Modality-Dependent Control in Diffusion CompositionDiffusion models → Medical imaging and segmentation
  35. Explainable Molecular Structure Inference from GC--MS with Diffusion Models and LLM RerankingDiffusion models → Flow matching, time series and science
  36. MOF-VERIFY: A Failure-Aware Agentic Harness for MOF Hypothesis VerificationLLM reasoning → Reasoning for agents, planning and code +1
  37. hacktrace: behavior-supervised detection of reward hacking during code generationLLM agents and tool use → Software engineering coding agents +1
  38. Parasitic Co-Denoising: Unlocking 3D Human Motion Generation in a Frozen Video Diffusion ModelDiffusion models → Video generation
  39. HyperThink: Text-to-Parameter Hypernetworks for Efficient ReasoningLLM reasoning → Test-time compute and efficient thinking
  40. WebFovea: When the Model Is Right but the Click Is Wrong -- Reliable Round Trips for Vision-Based Web Agents on Live WebsitesLLM agents and tool use → Multimodal and embodied agents
  41. Adaptive Second-Order Solvers for Fast Stochastic Diffusion SamplingDiffusion models → Diffusion foundations and sampling
  42. Tailoring the Quantization Space for 1-Bit KV Cache CompressionEfficient LLM inference and serving → KV cache quantization
  43. DyadMem: A Long-Term Memory Benchmark of How Agents Work with UsersLLM agents and tool use → Agent memory systems
  44. Beyond Predefined Sinks: Security-Aware Dependency Analysis for LLM AgentsLLM agents and tool use → Agent threats, guardrails and governance
  45. AvoKV-E: Payload-Aware KV Cache Eviction for Long ReasoningEfficient LLM inference and serving → Efficient reasoning and agent inference
  46. Sentry: Learning to Recover from LLM Agent Failures at Test TimeLLM agents and tool use → Agent skill learning and evolution
  47. OLMo-Detect: A Multi-Stage, Confounder-Controlled Benchmark for Membership Inference on Large Language ModelsLLM safety and security → Membership inference attacks
  48. A Guideline-Augmented Multi-Agent Framework for Schema-as-Code Biomedical Named Entity RecognitionLLM agents and tool use → Medical and scientific domain agents
  49. Reasoning with Evidence, Not Merely Rationales: Verifiable Preference Proofs for LLM-Based RecommendationLLM reasoning → LLM reasoning for recommendation
  50. Post-Training Frontier Text-to-Image Models by Composing Preference and Rubric RewardsLLM alignment / RLHF / preference optimisation → Diffusion and multimodal model alignment
  51. Understanding Trajectory Heterogeneity in Federated World Model LearningWorld models → Medical and surgical world models
  52. SlimKV: Joint Token-Feature KV Cache Compression with Reconstruction-Free Beacon AttentionEfficient LLM inference and serving → KV cache optimization
  53. GTDD: Generative Test-Driven Development for AI Coding Agents with Adversarial TestingLLM agents and tool use → Software engineering coding agents
  54. Dynamic Expert Pruning for Multi-Agent SystemsLLM agents and tool use → General LLM agent frameworks and benchmarks +1
  55. Continual Graph Memory for Mathematical Research AgentsLLM agents and tool use → Agent memory systems
  56. When to Compile a Computer-Use Agent? Measuring Payback and Making Compilation Decisions for Token EfficiencyLLM agents and tool use → Software engineering coding agents
  57. Positive-Unlabeled Learning for Agent Safety False Alarm AuditingLLM agents and tool use → Agent threats, guardrails and governance
  58. Frequency Is Not Sensitivity Identifying Safety-Sensitive Experts in Sparse MoE LLMMixture-of-experts → Expert routing strategies
  59. MixVLA: Adaptive Mixing of Non-Invariant Information for Generalizable Vision-Language-Action ModelsVision-language-action models → Generalist VLA models and benchmarks
  60. Revealing Epistemic Uncertainty in MLLMs via Causal-Invariant MaskingMultimodal LLMs / vision-language models → MLLM training and visual instruction tuning
  61. PsyEvo: A Personalized Counseling Agent That Self-Evolves at Test TimeLLM agents and tool use → Social simulation and multi-agent debate +1
  62. Found but Not Read: When Extracted Text Closes the Retrieval-Reading Gap in Document Vision-Language ModelsMultimodal LLMs / vision-language models → Document understanding and multimodal RAG +1
  63. Evaluating VQA in Vision Language Models using Cooperative PrinciplesMultimodal LLMs / vision-language models → VLM benchmarks and visual reasoning
  64. Seeing, Saying, but Not Using: From Reportable Spatial Facts to Usable States in Multimodal Large Language ModelsMultimodal LLMs / vision-language models → Embodied and 3D vision-language models
  65. On Unlearning for Time-series ForecastingLLM safety and security → Machine unlearning
  66. Counterfactual Action Evaluation, Observation Bottlenecks, and Representation Geometry in Joint-Embedding Predictive World ModelsWorld models → Latent predictive world models
  67. Harness-Aware Distillation for Small Language Model AgentsLLM agents and tool use → General LLM agent frameworks and benchmarks +1
  68. Bounded Reachability & Jailbreak Detection via Contraction-Constrained State Space ModelsLLM safety and security → Safety alignment, refusal and guardrails
  69. Understanding Enrichment in Reinforcement LearningDeep reinforcement learning methods → Deep RL algorithms and representations
  70. DNAlign: Dynamic Null-Space Safe Alignment for LLMsLLM alignment / RLHF / preference optimisation → Safety alignment and jailbreak attacks +1
  71. PointWAM: 3D World Action Modeling for Dexterous Robotic ManipulationWorld models → Embodied and robot world models
  72. All Work And No Play Makes Jack a Dull Boy: Understanding and Preventing Catastrophic Strategy Collapse in RLVRLLM reasoning → Reinforcement learning for reasoning
  73. FastOPD: On-Policy Distillation for Lightweight VLA DeploymentVision-language-action models → Efficient VLA inference and adaptation +1
  74. AMBER: Multi-View Adaptive Budget Allocation for Listwise Vision-Language RerankingMultimodal LLMs / vision-language models → Document understanding and multimodal RAG
  75. Clinical Concept Centers in LLMsMechanistic interpretability → Knowledge and representations inside LLMs
  76. MLCommons Jailbreak Benchmark v1.0LLM safety and security → Jailbreaking
  77. MetaRubric: Learning to Reward for Rubric-Based Reinforcement LearningDeep reinforcement learning methods → RL training of language and agent models
  78. Text-Centric Post-Training for Omni-Modal ReasoningLLM reasoning → Multimodal reasoning
  79. RMCW: A Deletion-Robust Watermark Based on Reed--Muller Codes for Language ModelsLLM safety and security → Watermarking generated content and models
  80. iS-KV: Online Low-Rank KV Cache Compression via Block-Incremental SVDEfficient LLM inference and serving → KV cache optimization
  81. ManiPhysicsBench: Physics-Based Assessment of Object Preservation in VLA ManipulationVision-language-action models → Tactile, force-aware and humanoid VLA
  82. VIGOR: Zero-Shot Visual Generalization via Latent-Space Consistency in Model-Based Reinforcement LearningDeep reinforcement learning methods → RL with diffusion and flow models
  83. BitNest: Bit-Nested Speculative Decoding for Memory-Efficient LLM Inference AccelerationEfficient LLM inference and serving → Speculative decoding
  84. Skill2Real: Agentic Skill Learning for Zero-Shot Sim-to-Real Robot ManipulationLLM agents and tool use → Agent skill learning and evolution
  85. SimpleTouch: Can Vision-Language-Action Models Master Contact-Rich Manipulation Without Tactile Policy Pretraining?Vision-language-action models → Tactile, force-aware and humanoid VLA
  86. OPD Before RL: Warm-Starting Rubric-Based RL with On-Policy DistillationKnowledge distillation → On-policy and self-distillation for LLMs
  87. Controlling Polar Exposure to Delay Memorization in Diffusion ModelsDiffusion models → Diffusion foundations and sampling
  88. Localized Conformal Safety Monitoring with Vision-Language Models for Autonomous DrivingMultimodal LLMs / vision-language models → Driving VLMs and open-vocabulary detection +1
  89. Jumping up and down: Denoiser diffusion models for discrete ordinal dataDiffusion models → Diffusion foundations and sampling
  90. Correcting Guided Diffusion Trajectories with Spectral AlignmentDiffusion models → Diffusion foundations and sampling
  91. EpiWorld: Grounding LLM Policy Agents in Epidemiological World ModelsWorld models → World models for language agents
  92. On the Chain-of-Thought Monitorability of Looped Language ModelsLLM reasoning → Prompting and chain-of-thought reasoning
  93. Prospective Hindsight: Self-Calibrating Reinforcement Learning via Prediction-Reality GapsDeep reinforcement learning methods → Deep RL algorithms and representations
  94. Beyond Correctness: Resolving Underspecification in Agentic Text-to-SQLLLM agents and tool use → Web, search, and research agents
  95. Bellman Error Minimization Via Linear Programming NormalizationDeep reinforcement learning methods → RL theory and sample complexity
  96. SymRegFlow: Symmetry-Regularized Flow Matching for Video World ModelsWorld models → Video-generation world models and simulators +1
  97. Structural-Functional Brain Connectivity Generation via Multimodal Hypergraph-based Flow MatchingDiffusion models → Medical imaging and segmentation
  98. Revisiting Visual Representation Enhancement of VLMs via Kernel Canonical Correlation AnalysisMultimodal LLMs / vision-language models → Prompt learning and adaptation of CLIP
  99. WakeKV: Reactive, Reversible KV Residency for Heads That Change Their MindsEfficient LLM inference and serving → KV cache optimization
  100. AdaTempo: Learning Shared Relative Tempo from Demonstrations for Faster Robot ManipulationOffline RL and imitation learning from data → Robot manipulation imitation learning
  101. Learning to Revise Reasoning with Segment-wise On-Policy DistillationLLM reasoning → Prompting and chain-of-thought reasoning +1
  102. Silent Dissent: LLM Agents That Yield to the Majority Still Represent Their Original PremiseLLM agents and tool use → Social simulation and multi-agent debate
  103. Conditional Capacity and Routing in Mixture-of-Experts Particle TransformersMixture-of-experts → Expert routing strategies
  104. Learning from Evolving Errors: Adaptive Iterative Repair for On-Policy DistillationLLM reasoning → Reinforcement learning for reasoning +2
  105. Test-time Calibration Learning for Large Language Model ReasoningLLM reasoning → Test-time compute and efficient thinking
  106. Spend Teacher Tokens Where They Matter: Success-Referenced On-Policy DistillationKnowledge distillation → On-policy and self-distillation for LLMs
  107. LEAP: Learning Efficient Action Proposals For LLM AgentsLLM agents and tool use → Reinforcement learning for LLM agents
  108. CHASE-VLA: Post-Training Quantization Framework for Vision-Language-Action Models with Chunk-Aware Scale EstimationVision-language-action models → Efficient VLA inference and adaptation
  109. Large Language Continuous Diffusion ModelsDiffusion models → Discrete and text diffusion +1
  110. A GHOST in Long-Horizon Agents: Governance Hazard from Overlooked Safety Constraints across TurnsLLM agents and tool use → Agent threats, guardrails and governance
  111. Generalization Bounds for Flow-matching Generative Models for Intrinsically Low-dimensional DataDiffusion models → Flow matching, time series and science
  112. SpectralCache: Accelerating Diffusion-Based World Models via Spectral Feature CachingWorld models → Video-generation world models and simulators
  113. Distributed Learning with Selective State Space Models: Architecture-Aware Convergence AnalysisState-space sequence models (S4, Mamba) → Core SSM architectures for language
  114. Context-Tower Conversion Preserves Generation While Freezing Retains Knowledge: Low-Budget AR-to-Diffusion Conversion of MoE LLMsDiffusion models → Discrete and text diffusion +1
  115. Coherence-Driven Belief Formation and Population Dynamics of Contagion in LLM AgentsLLM agents and tool use → Social simulation and multi-agent debate
  116. Equivariant Flow Matching for Electron Density PredictionDiffusion models → Flow matching, time series and science
  117. Imagine the Future, Internalize the Gist: Efficient VLA Reasoning via Internalized Spatiotemporal ImaginationVision-language-action models → World model-based VLA
  118. CuBEs: Culturally-Situated Behavioral Evaluations and the Limitations of Culture-Blind LLM JudgesWorld models → World models for language agents
  119. WebUIProof: Benchmarking WebUI Code Generators with UI-Agent Execution HarnessLLM agents and tool use → GUI and mobile agents
  120. VERSE: Verified Self-Evolving Optimizer for Agent HarnessesLLM agents and tool use → Software engineering coding agents +1
  121. Scale-Recursive Rectified Flows for Few-Step Precipitation EnsemblesDiffusion models → Flow matching, time series and science
  122. KNOU-Flood: Advancing global streamflow forecasting with lead-aware probabilistic machine learning
  123. Same Answer, New Box: How Much of RLVR's Out-of-Domain Gain Is Answer Format?
  124. Expanding Attention: A Taylor-Series Route to Linear-Time Transformers with a Fixed Number of Recurrent States
  125. Does Dynamic-Point Filtering Help When Texture Is Scarce? A Controlled Study of ORB-SLAM2 Front-Ends in Synthetic Indoor Scenes
  126. Abdominal Ultrasound Simulation from Semantic Labels using Paired Label-to-Physics-Based Image Translation
  127. STRIDE: Spatial-Temporal Representation for Interval-conditioned Disease Evolution in Longitudinal Glioblastoma MRI
  128. Extending Music Annotation Schemas: Zero-Shot Prediction or Few-Shot Adaptation?
  129. Anchor Divergence for Semantic Geometry in Contrastive Learning
  130. A Governed Agent Architecture for Natural-Language Dashboard Generation over Enterprise DataBrain
  131. One-Cycle Fault Classification and Faulted-Line Identification on the PROTECT-90 Dataset: An Initial Application Benchmark
  132. Watermarks and Fingerprints as Soft Bindings for Content Provenance: An Open-Licence Benchmark for Images, Audio and Video
  133. You May Be Running the Wrong Inception Crop
  134. Consideration Circuits: Depth Separation and Universality Beyond a Single Softmax
  135. Ideal Paths for Approximating Logistic Gradient Descent Trajectories at Large Initialization
  136. Dual-Scale Relational Graph Transformers for Ecosystem-Aware Fraud Detection
  137. Agentic Resource Allocation for Batch Multi-Objective Bayesian Optimization in Autonomous Materials Discovery
  138. Shared Geometry Is Not Shared Physics: A Layerwise Test of the Platonic Representation Hypothesis in Astronomy
  139. What Gradients Add to Text Leakage in Split Language Models, Counted per Token and per Document
  140. Representational Control over Self-Report & Behavior Coherence in LLM Risk-Taking
  141. Agent Reliability Profiles in Financial Services
  142. LongSocialBench: Do Long-Context LLMs Understand Online Discussion Threads?
  143. Sharp Convergence and Sample Complexity of Policy Mirror Descent for Average-Reward MDPs
  144. Physics is the Best Teacher: Consistency Learning for Time-Invariant Operators of Chaotic Dynamics
  145. Where Does the Semantic Gain Come From? A Reproduction and Extension of Semantic Knowledge-driven Contrastive Learning for Long-Tailed Recognition
  146. Progressive Multi-Ancestor Bit-Depth Distillation
  147. Scaling 3D Visual Grounding in Abdominal CT
  148. Beyond Masked Sparsity: SNACK Enables Truly Sparse Neural Networks on GPU
  149. UniBRep: Learning Unified Geometry and Topology for Image-conditioned B-Rep Generation
  150. Robust blind unmixing: A geometric approach to overcoming basis variation
  151. Exact Optimal Transport by Matching
  152. Dependable AI-Assisted Engineering: A Formal Framework for AI Participation and Assurance in Safety-Critical Workflows
  153. AEGIS: Differentiable Mars Climate Model with Neural Closures
  154. A study on human-agent teaming through spoken interaction: the impact of human individual traits and agent characteristics
  155. VCURF: Virtual Camera-based Uncertainty of Radiance Fields
  156. Why Convolution Still Matters: Evaluating Inductive Biases in Cryospheric Image Classification
  157. Articulatory Entrainment and Coordination Complexity in Spontaneous Autistic and Non-autistic Dialogue
  158. On architectural choices for interpretability and thermodynamic consistency in Physically Recurrent Neural Networks in the low-data regime
  159. Evaluating Zone-Guided Front Extraction for Glacier Calving-Front Delineation in SAR Imagery
  160. Application of sequence learning for predicting radiation damage of the CMS electromagnetic calorimeter
  161. The Cost of a Hop: Benchmarking NLIP and A2A
  162. A Theory of Shape Reconstruction from Heat Conduction and Shading
  163. ASD-FEAT: A Multi-Modal Infant Video-Derived Dataset for Early ASD Risk Prediction
  164. Geometry-Dependent Approximation for Non-Monotone k-Submodular Maximization
  165. Periscope: Extending Frozen Language Models Beyond Their Context Window
  166. Dynamic Quadtree Tokenization and Transformer for Adaptive Mesh PDE Forecasting
  167. Adaptive Partitioning Schemes for Optimistic Optimization
  168. MAGEFormer: Learning Metric-Consistent Representations for Anisotropic CT Segmentation
  169. LD-EnFF: Latent-Dynamics Ensemble Flow Filtering for Data Assimilation with Sparse Observations
  170. Gated Graph Neural Networks for Learning Hidden Independent Cascade Dynamics
  171. Learning Subject-Specific Anatomical Representations via Manifold Expansion: Application to Accelerated Multi-Contrast MRI
  172. The Reported Engagement with AI Level (REAL) Rating: A Framework for Disclosing Human-AI Collaboration
  173. A Quantitative Analysis of Graph Representation Strategies for Cyber Attack Detection
  174. Beyond the Parameter Monolith: Reconstructive Memories, Executable Skills, and Residual Assembly for Language Models
  175. Exploration-Preserving Policy Optimization
  176. VolS-GS: Relightable Gaussian Splatting with Volumetric Subsurface Scattering
  177. Verifier-Guided Synthetic Augmentation for 3D Human Shape Generation
  178. Selective Backpropagation for Efficient Few-Shot Class-Incremental Learning
  179. Do Language Models Need a Trainable Input Embedding Table? Fixed Minimal Token Codes at 1.7B-Class Scale
  180. Behavioral History Outperforms Descriptions of the Person for LLM Synthetic Personas
  181. Solving VeriContest with a Lean-Backed Rust Verifier
  182. Masked Privileged-Information Distillation for Multimodal Skin Lesion Classification Under Missing Clinical Metadata
  183. tildeO(√T) Regret and Polylogarithmic Constraint Violation for COCO
  184. Learning Latent Protein Languages for Autoregressive Generation
  185. Probabilistic Algorithms for Ising Machines from Optimization to Generative AI
  186. An Executable Benchmark for LLM-Based HLS Repair:Design Complexity and Repair Underconstraint
  187. ROAR: Unifying Runs across Heterogeneous AI-Driven Research Systems
  188. BAIBAICHUCHU at the NTCIR-19 FinArg-3 Task: When Is Maximum Possible Profit Predictable from Investor Text?
  189. Localize-and-Detect: Auditing Task-Level Poisoning in Instruction-Tuned Models
  190. Latent Score-Based Bayesian Cram'er-Rao Bound Estimation for High-Dimensional Imaging Systems
  191. One-Step Curvature Probes Miss the Fitting Operator: Retained Capacity and Terminal Null-Space Correction for Continual Learning
  192. Evolving LLM-Generated Features for Interpretable Classification
  193. Synthesizing Physics Formulae with Transformers
  194. DePICT: Decision-Preserving Interface for Constrained Downstream Tasks
  195. A Step Towards Forgetting: Optimiser History and the Loss of Answer Mass
  196. TreeWalker: Partial Evaluation for Grouped Tree-Ensemble Inference
  197. General Decision Models: Benchmarking and Insights Beyond Jev
  198. SGAnalog: An End-to-End Circuit Benchmark from Open-Source Silicon Tapeouts
  199. DABACO: A Multi-Camera Dataset and Benchmark for Screen Localization and Pointing Estimation
  200. Co-Supervised Tree Synthesis for Interpretable Subgroup Identification and Honest Treatment Effect Inference in Randomized Trials
  201. COVER: Learning to Accept More in Selective Sleep Staging
  202. AdaEva: Accelerating LLM-Driven Algorithm Design with Adaptive Partial Evaluation
  203. REACT: Physically and Chemically Consistent Reconstruction of Marine Active Tracers
  204. Streaming Multi-Track Timeline Control for 3D Human Motion Generation
  205. Training Numerical Intelligence via Auto-Diagnosis and Skill Discovery
  206. GOTT: Object-centric Dexterous Manipulation with a Reusable Cross-Embodiment Primitive
  207. Retrieval-Centric Deep Learning in Growing Nonparametric Neural Networks
  208. BAT-NO: A Boundary-Condition-Aware Transformer Neural Operator for Crashworthiness Prediction of Vehicle Components
  209. LLM-enhanced spatio-temporal learning for grid-level docked bike sharing demand prediction
  210. Distribution Matching Evolutionary Algorithms for Rare Event Sampling
  211. Where Does Jagged Competence Come From?
  212. OncoNoteBERT: A Foundation Representation Model for Natural Language Processing of Real-World Outpatient Oncology Notes
  213. The Score Is Not the Structure: Brain Alignment and Cross-Lingual Transfer
  214. Are We Measuring Anticipation? Auditing Privileged Information in Procedural Video Evaluation
  215. Same Output, Different Gold: Measuring How Reference Choice Moves a Multilingual Benchmark Score
  216. Targeted Active Learning for Preference-Based Treatment Effects on Multivariate Outcomes
  217. Beyond Token Accuracy: Prioritizing What Matters for Visual Reconstruction
  218. Less Decoder is More Encoder: Geometric Representation Learning from Novel View Synthesis
  219. MoSE3: Learning World-Space SE(3) at Every Pixel
  220. 4DCodeBench: Benchmarking Agents on Inverse Graphics of Dynamic Scenes
  221. RNADyn: A Benchmark for Generating and Understanding RNA Dynamics
  222. EyeRobot 2.0: Active Gaze for Precise Manipulation without Wrist Cameras
  223. From Mixing to Tearing: Graph Decomposition in Decentralized Optimization via Message Passing
  224. LESSER: Post-Training Data Selection with Output-Layer Gradients
  225. Language Models that Play Chess and Explain Their Moves
  226. Transcriptome-informed multi-modal AI for predicting neoadjuvant therapy response from breast cancer biopsies
  227. FlowHMR: Physically Plausible Motion Capture from Video
  228. SigLIP2 for aerial fire risk classification
  229. Simulation-Free Learning of Population Dynamics with Wasserstein Lagrangian Residuals
  230. FrugalEvo: Towards Cost-Aware LLM-Guided Program Evolution
  231. Planning to Learn
  232. ProAR: Learning Prospective Reasoning with Autoregressive Video Models
  233. Forecasting from Counterfactual Simulator Rollouts: A Sim2Real Evaluation
  234. Revisiting Input Time-frequency Representations in Multi-pitch Estimation for Vocal Ensembles
  235. MRVQ: One Resident Index for Dimension- and Rate-Elastic Vector Search
  236. PoCoFL: POlicy-COmpliant Federated Learning
  237. On-Board Anomaly Detection for Efficient Marine Environmental Monitoring
  238. Amortized Structured Stochastic Variational Inference for Gaussian Process Latent Variable Models
  239. When May a Bandit Leave Its Anchor? E-Process-Authorized Thompson Sampling under Non-stationarity
  240. On the Convergence of Success Conditioning for Policy Optimization
  241. Broken scale symmetries in undercomplete linear autoencoders
  242. Do Large Language Models Know Colombian Law? A Reliability Benchmark for the Colombian Legal System
  243. LoGo: Local-Global Rewards for Consistent Long-Horizon Video Generation
  244. Depth as Time in One-Step Generative Models
  245. FALCON: A Model and Dataset Agnostic Framework for Synthetic Data Generation for NL2SQL Pairs
  246. Normal-Form Correlation in Markov Games
  247. Video-Time Priors for EEG-fNIRS Emotion Regression on Familiar Videos
  248. ManifoldSplat: Language-Guided Semantic Shape Editing of 3D Gaussian Head Avatars
  249. A Path Integral Surrogate for Multi-Step Gradient Inversion in Federated Learning
  250. HyperBrowseComp: A Multilingual and Multimodal Stress Test for Web-Browsing Agents
  251. Learning to Assess Heartbeat Observability for mmWave Heart-Rate Sensing
  252. Writerslogic at the CLEF 2026 SimpleText Track: Multi-Candidate LLM Simplification and Stacked Complexity Spotting
  253. Knowledge or Calculator? Decomposing the Skill Premium in Verifiable Financial Agent Workflows
  254. Cephalonauts One: A deep fMRI dataset for decoding naturalistic speech in the human brain
  255. Get a GRIP, this will be a long TRIP: A Quantifiable Long-Range Framework for Verifying Over-squashing
  256. Objects Without Morphisms: What LLMs for Mathematics Do Not Represent
  257. Recursive Harness Self-Improvement for Frontier Reasoning Data Synthesis
  258. ZeroMAG: Zero-Shot Multimodal Adapter Generation for Plug-and-Play EEG Foundation Models
  259. DuoMatching: Joint-Marginal Distribution Matching for Few-Step Video Generation
  260. Autonomous Robotic Navigation for Endovascular Brain-Computer Interface Access
  261. Author Representation Strategies for Zero-Shot Authorship Attribution: A Comparative Study of LLM-Based and Embedding-Based Approaches
  262. Beyond Trained Models: Compiling GNNs for a Sound Explainer Benchmark
  263. Structured Composition of Verifiable Atomic Insights for Table-to-Report Generation
  264. From Benchmarks to Production: A Text-to-SQL System for Complex Financial Data
  265. Feedforward Novel View Synthesis for Heterogeneous Cameras
  266. ProgressNet: Sketching and Prompting with a Frozen Text-to-Image Model
  267. Weave Forcing: Compositional Memory Routing for Interactive Long Video Generation
  268. An Automated and Reproducible Workflow for Crack Identification and Damage Assessment of Fusion Materials
  269. Below what training size do deep tabular generators stop beating trivial baselines? A preregistered benchmark on a size ladder of clinical and standard datasets
  270. Most-Recent Anchoring with Recurrent Ordering for Time Series Forecasting
  271. Dual-Context Analog Retrieval for Time Series Forecasting
  272. Single-Pass Uncertainty Heads for Claim-Level Hallucination Detection in Persian Medical Language Models
  273. Metropolis-Hastings Dominates Importance Resampling for Policy Composition
  274. Fed-ADApt: Federated Anytime Depth Adaptation for Resource-Aware Medical Image Segmentation
  275. UniDynamics: Event-RGB Fusion for Unified Future 4D Dynamic Scene Generation
  276. Preserving Anatomical Continuity: Three-Stage Pipeline for Colon Segmentation in 3D Abdominal CT Scans
  277. When Is Accuracy Evidence? A Unified Theory of Generalisation, Validation, and Information Fusion
  278. Generalization of Transformer-Based Neural Quantum States via In-Context Learning
  279. Beyond Random Splits: Evaluating Drug-Target Affinity Models Under Chemically and Biologically Motivated Distribution Shifts Copy
  280. Measure Less, Know More: Self-Supervised Test-Time Feature Acquisition
  281. I2CD: Direct Image-to-Convex Decomposition for Simulation-Ready Collision Geometry
  282. Causal Representation Learning with Instantaneous and Lagged Relations via Nonstationarity
  283. Electronic Density versus Geometry for Machine-Learned Molecular Absorption Spectra
  284. ChromaGS: Text-Driven Semantic Editing of 4D Gaussian Avatars
  285. Depth Hypothesis Guided Iterative Refinement for Event-Image Monocular Depth Estimation
  286. The Shape of Speech: A Geometric Measure of Coarticulation for Speech-Driven 3D Facial Animation
  287. OptiSelect: How does the Optimizer Shape Data Curriculum?
  288. Jumping the Line: Exploiting Length Predictions in LLM Scheduling
  289. Becoming Suspicious Across Borders: Algorithmic Extraterritoriality and AI-Driven Financial Surveillance
  290. Deep Bayesian REFoCUS
  291. Rethinking Epistemic Uncertainty in Node Classification through Information Growth
  292. AIBL: Augmented Instance-Based Learning with Structured Memory and Neural Embeddings
  293. Contrastive Neural Embeddings Reveal Individual Traits Beyond Conversational Role
  294. ForestQuery: Boundary-Aware and Spatially Anchored Query Learning for Unified Forest Point Cloud Segmentation
  295. 16-bit Precision of Convolutional Neural Networks on Microcontroller Units for 8-bit Costs
  296. Benchmarking candidate coverage and rejection policy transfer in typed decision models
  297. CVE2AP: Automated Generation of PDDL-Encoded Attack Paths via Large Language Models
  298. Interpretable Deepfake Detection in Videos via Explicit Forensic Features and Temporal Modeling
  299. Operator-informed initialization for Fourier features physics-informed neural networks
  300. SCAD: Structured Credit Assignment and Distillation for Long-Horizon Agents
  301. LAS-CLIP: A Lightweight Adapter Steering Approach for CLIP's Visual Encoder
  302. Multilingual GSM-Symbolic: What determines capability transfer across languages?
  303. Geometry Meets Physics: Data-Efficient Pre-Training for Unstructured Neural PDE Solvers
  304. A Fully Automatic Pipeline for 3D Dendrite Instance Segmentation in SBF-SEM
  305. Cordial Learning: Distributed Training with Correlated Data
  306. To Jev or Not? Evaluating the Accuracy and Efficiency of Structured Decision Models for Hate-Speech Moderation
  307. Information Limits of Low-Rank Approximation Certification
  308. Multi-Task Evolution for Zero-Shot Cross-Problem Generalization using LLMs
  309. SDECast: Probabilistic Weather Forecasting in Continuous Time with Neural SDEs
  310. Optimal Planning in a Dynamic World
  311. T3lescope: Arbitrary-Resolution High-Fidelity Generative Surface Reconstruction from Images
  312. Training-Loss Guarantees for Muon with Finite-Step Newton--Schulz Orthogonalization
  313. S2-PINN: Stochastic Separable Physics-Informed Neural Networks
  314. JOVE: Joint Execution and Verification for Resource-Aware LLM Task Graphs
  315. Wrong Organ, Right Physics: Transferring Echocardiography Pretraining to Lung Ultrasound for Tuberculosis Screening
  316. HexVIO: Towards All-Day Stereo-Inertial Tracking Through Commodity DSPs
  317. Moving Forward with Video Saliency: A New Dataset and Benchmark where Motion Matters
  318. EVOL: Simulator-Guided Evolutionary Expert Synthesis for Deployment-Free Learning Path Recommendation
  319. Shrome at Touch'e: Soft-Vote Ensembling and Counter-Causal Augmentation for Causality Extraction
  320. PaMIR: Open Benchmark of Public Credit-Default Datasets
  321. Mapping and Advancing the Scalability-Accuracy Frontier of Nonlinear Causal Discovery
  322. Cross-cohort TB classification using clinical data gathered in Uganda and South Africa
  323. COSMI: COmpositional Synthesis of Multi-object Interactions
  324. Learning a Fact Is Not Learning How to Retrieve It
  325. EmbPASS: Towards Cross-Embodiment Open Panoramic Segmentation
  326. Collective Bias Mitigation via Model Routing and Collaboration
  327. WAMpy: Efficient Synthesis of Prolog Programs in Python
  328. The Neuro-Physical Inverter: A Modular Framework for Magnetotelluric Inversion Coupling Ensemble Conditioning with Residual Learning
  329. Near-Optimal Convex Optimization with Lazy Second-Order Oracles
  330. Evolving Hybrid Quantum-Classical Architectures for Image Classification
  331. VisionMX: Unlocking Microscaling Post-Training Quantization for Vision Models
  332. Kernel Singular Value Decomposition with Extension to Multiple Data Sources
  333. StanceEval 2026: The Second Stance Detection Shared Task
  334. HyperFuse: Fast Self-Supervised Node Embeddings for Attributed Hypergraphs
  335. Hamiltonian locality testing and certification do not achieve the Heisenberg limit
  336. Predictively Oriented Gaussian Process Posteriors
  337. Predicting and Repairing Merge Collapse in Large Language Models
  338. Bridging Research and Practice: A Systematic Evaluation of Generalist and Dermatology-Specific Models in Clinical Skin Lesion Classification
  339. PocketSplat: Mobile Gaussian Reconstruction via World-Space Latent Allocatio
  340. Not Until the Evidence Says So: Teaching LLM Investigators When to Close a Case
  341. Lightweight and Resource-Efficient Perception for Robotic Guide Dogs
  342. Geometry-Aligned Semantic Matching for Cross-Modal Planar Image Registration
  343. LiBRA: Detection-Aware Image Watermark Removal via Bidirectional Latent Optimization
  344. Budgeted-GS: Real-Time Large-Scale Gaussian Splatting via Factoring LOD
  345. Landscape-Dependent Performance of Photonic Quantum Solvers in QUBO Feature Selection for Financial Risk Detection
  346. Multimodal reasoning for broadly neutralizing antibody discovery from label-free human B cell repertoires across virus families
  347. TSGuard: A Real-Time Framework for Detecting and Imputing Missing Data in Streaming Time Series
  348. CalCErt: Bin-wise Certification of Confidence Calibration in Medical Image Classification
  349. Benchmarking Literature Retrieval for a Model Organism: A Dictyostelium Case Study
  350. Trading Strategy Optimization via Textual Gradient
  351. Coverage You Can Steer: Online Conformal Calibration for RL-Driven Hardware-Aware NAS
  352. ParaGeo: Decomposing Paralinguistic Variation into a Shared Latent Geometry
  353. Exploring the Trade-Off Between Structured Pruning and Fault Tolerance in Deep Neural Networks for Space Applications
  354. Building Interpretable Feature Representations for Resume-Vacancy Matching by Distilling Production LLM Signals
  355. Ontological Instability and Statistical Amplification: The Paradox of "Humanizing" LLM-Generated Text
  356. Emergent Structure in the Marginal Attention Space of Language Models
  357. S2S-JEPA: Predicting the Predictable at Subseasonal-to-Seasonal Timescales
  358. A Benchmark for Spatially Grounded Gesture Generation
  359. Invariance of Clustering Operations in Causal Effect Identification
  360. Beyond Single Videos: Benchmarking and Active Evidence Seeking for E-Commerce Cross-Video Reasoning
  361. Predictor-Guided Latent Space Codon Optimization for Maximizing Protein Expression
  362. Peer Influence across Heterogeneous AI Models
  363. LS-AR: Future-Predictive Latent Steering in Autoregressive LLMs
  364. ULTRADISCOVERY: Abductive Exploration in an Interconnected, Epistemically Open Universe
  365. Zephon: Elastic Determinism for Online, Stateful Foundation Model Data Loading Pipelines
  366. Light Entropic Optimal Transport on Riemannian Manifolds
  367. NegT2IBench: When Negation Changes the Picture. A Polarity Benchmark for Text-to-Image Models
  368. Smart Sensing for Safer Bridges: From Sensor Signals to AI-Driven Anomaly Detection
  369. MintEval: Do LLMs Implement the Trading Strategy You Asked For? A Behavioural-Equivalence Benchmark for Natural-Language-to-Strategy Code
  370. RIFAR: Reliability and Forgetting-Aware Replay for Continual Robot Learning
  371. An automated pipeline for standardised speech-unit annotation in spontaneous dialogue
  372. Unmasking Propaganda: A Comparative Analysis of Masked and Causal Language Models
  373. Shapley-based Structural Analysis of Neural Calibration for Stochastic Volatility Models
  374. SecJev: Bringing Security Expertise to System One Decision Models
  375. Learning Transferable Policies from Action-free Time Series Through Dynamical Embeddings
  376. HARPO: Hallucination-Aware Reinforcement Learning for Faithful and Creative Language Generation
  377. When Does Synthetic Relational Data Teach Models to Use Relations? Tracing Predictive Structure from Pretraining Data to Model Behavior
  378. RIPPLE in Still Water: Zero-Shot Clustering in Federated Learning with Wavelet Scattering Transform
  379. The Geometry of Knowledge Accessibility in Large Language Models
  380. BeeWhere: Segmenting Bumble Bee Colonies to Quantify Behavioral Effects
  381. Balancing Multimodal Learning via Functional Progress
  382. When Numbers Start Talking: Numerical Signalling and Strategic Behaviour Among LLMs
  383. CrowdOcc: Monocular Semantic Scene Completion for Quadruped Robots in Crowded Indoor Environments
  384. SoftGene: Protein Language Model-Enhanced Soft Prompting for Interpretable Gene Set Annotation
  385. Verifiable, Articulable, and Tacit Components of Preference
  386. ReSCUE: Re-translation with Sentence Commitment for Unsegmented Long-Form Simultaneous Sign Language Translation
  387. Signal Simplification Is Not Predictive Simplification: Diagnosing Residual Neural Forecasting in Short-Horizon Volatility
  388. Personalized Automatic Speech Recognition for a Dysarthric and Tracheostomic Speaker using Artificial Conversations
  389. From Expression to Reaction: Role-aware Visual Transfer and Stimulus-guided Reasoning for Interlocutor Emotion Recognition
  390. OmniAct3D: Leveraging Foundation Geometry and Evidence-Grounded Reasoning for Panoramic 3D Detection
  391. RYOPO: Bringing End-to-End Category-Level Object Pose Estimation into Real Time
  392. Rethinking Fixed Temporal Grids: Frequency-Disentangled Motion Generation
  393. Response Variability and Stability in Human Reasoning
  394. Recursive Self-Improvement in Unified Multimodal Models
  395. Neural Data Needs Semantic Tokenization: Behavioral Events as Boundaries of Session-Transferable Tokens
  396. Temporal Geometry of Deep Networks: Hyperbolic Representations of Training Dynamics for Intrinsic Explainability
  397. OmniConfess: Eliciting Token Confessions to Mitigate Omni-Modal Hallucination
  398. Differentiable Koopman Operator for Contrastive Learning on Dynamic Graphs
  399. PLCWorld: Benchmarking LLM-Generated PLC Programs in Closed-Loop Plant Simulation
  400. Safeguarding Mutual Correction in Source-Free Domain Adaptation via Cut Statistics
  401. RASPER: Reward-Aligned Summarization of Clinical Notes for EHR Outcome Prediction
  402. Relevant Evidence Decoding for Audio-Visual Hallucination Mitigation
  403. Reliable Self-Evolution with Imperfect Proxy Rewards
  404. From Language Priors to Field Adaptation: Preference Learning for Traversability Estimation
  405. CreateScore: Domain-Theory-Informed Bayesian Routing for LLM-Based CV Screening
  406. Dirac-Interconnected Neural Elements: Discovering Modularity in Physical Systems Without Reduction
  407. TerraVis: Towards Evaluation of World-Grounded Visual Consistency in Text-to-Image Generation via MLLM Workflows
  408. Hyperparameter selection for equation learning with biologically-informed neural networks
  409. Enhancing Biomedical Named Entity Recognition via Multiple Programming Languages Instruction Tuning and Ensemble Method
  410. When Predicting Nothing Beats SAM 3: Revisiting Evaluation in Video Object Segmentation
  411. Cross-Fitting Under Nonregularity: Normality and Inference via Locality
  412. Kinematics-Induced Multimodal 3D Human Pose Estimation with Subject-Level Privacy
  413. FASTDIAR: Frame-level speaker encoder for Streaming Diarization
  414. Discriminating Fixture Coverage in Agent-Infrastructure Verification Suites
  415. Output Language Confusion under Multilingual Prompt Contamination
  416. HASTE: Evolving Agent Harnesses Against Emerging Attacks Using Sparse Evidence
  417. Learning Jazz Pianist Style with Cross-Attention Conditioning
  418. Custom Forcing: Training-Free Subject Customization for Autoregressive Video Generation
  419. Probe the Harness: Setup Checks for Stale-Data RL Comparisons in Language Models
  420. Constraint-Aware Training
  421. Do ResNets Route? Sparse Interaction Experts in Residual Networks
  422. Tangent Schr"odinger Bridge Matching: Learning Stochastic Transport with Mechanistic Sensitivities
  423. ViTok: Improving Dense Semantics in AM-RADIO-Style Multi-Teacher Distillation with PHI-S and Masked Image Modelling
  424. LUMOS: Tracing Parametric Knowledge from Training Data to Behavioral Outputs in LLMs
  425. Interpreting at Write Time: A Policy Ablation for Multi-Goal Agent Memory
  426. When Can We Trust the Matching Principle? Robust Deployment Geometry Under Finite-Sample and Model Uncertainty
  427. A Residual Tree Gaussian Process Modeling Framework for High-Dimensional Data
  428. Misinformation Without Triggers: From Factual Answers to Downstream Decisions
  429. DyRA: Dynamic Residual Approximation for Efficient Matrix Multiplication in DNNs
  430. Toward Omni Multimodal Graph Foundation Model: A Topology-Driven Binding Approach
  431. Evaluating LLM-as-a-Judge Beyond Score Alignment: A Psychometric Analysis of Residual Judging Difficulty
  432. Query-aware routing for Cross-lingual performance gains in Encoders
  433. ConvoDrift: A Multi-Turn Conversational Dataset for Modeling Stylistic Tone Evolution
  434. Peer Effects in Signed Networks: Separating Influence Through Positive and Negative Ties
  435. AgentTrap: Stateful Feedback Deception against Autonomous Penetration Testing Agents
  436. Distributionally Robust Survival Models under Subpopulation Shift and Outlier Contamination
  437. TACD: Distilling Efficient Text-to-Motion Models via Terminal Amplification Control
  438. DIVINE: Simple Cross-Market Stock Pretraining via Diverse Indicator Reconstruction
  439. NeuroLens: Learning Latent Embeddings of Neural Semantics from Chronic Recordings
  440. Adaptive Mutual Distillation for Balanced Multi-Task Post-Training of Large Language Models
  441. Permutation Robustness Is Not Enough: Action Collapse in Multi-Agent Transformer Policies
  442. Turnover-Orthogonal Credit Assignment for Open-Team Multi-Agent Reinforcement Learning
  443. How Robust Is Multimodal Claim Verification to LLM Rewriting?
  444. To Explore The Strange New World Beyond Data Distribution: System Behavior, Causality Tax, and Non-causal Base Model
  445. FSPO: Policy-Consistent Risk and Pareto-Feasible Control for Budgeted LLM RL Post-Training
  446. Scaling Trajectories for Complex Tasks through Recursive Self-Rewrite
  447. TerrainForge: Physics-Grounded Road Geometry Editing for Counterfactual Autonomous Driving
  448. Adaptive Spectral-Koopman Dynamics Modeling for Temporal Domain Generalization
  449. Gated Slot Attention-2: Two-Sided Associative Memory Correction in Linear Attention
  450. ROUTEAUDIT: Interaction-Aware Identification for Budgeted Multi-Verifier Routing
  451. FUSEye: Training-Light Fisheye Detection with Overlapping Views and Zero-Initialized Adapters
  452. Muon Learns Facts Better: Understanding the Role of Spectral Orthogonalization
  453. Modeling Shared and Individual Structure for Cross-Subject Continuous Affect Regression from EEG-fNIRS
  454. Efficient Memory Crystallization for Graph Learning under Non-Stationary Distribution Shifts
  455. PAPER2LLM++: Continual Self-Evolution of LLMs from Research Papers
  456. Law And Order: Tax Law Autoformalization
  457. Hold-Out Scoring for Efficient Gaussian DAG Learning
  458. TRAC: Trajectory-aware Reuse and Adaptive Correction for Efficient Autoregressive Video Generation
  459. Automatic Evaluation of Mental Health Stigma in Online Communication
  460. LatticeSMC: Where to Spend Inference-Time Compute in Chunked Sequence Generators
  461. Improving Atomic-Fact Recall via Focused Views in Unstructured Knowledge Editing
  462. Nearly Optimal Fixed-Confidence Best-Arm Identification with 1-Bit Feedback
  463. AptMQL-Bench: From Text-to-SQL to Text-to-MQL via Access-Pattern Schema Design and Data-Preserving Migration
  464. When History Fails to Become Experience: Action Calibration in Language Agents
  465. No-Free-Graph: Learning When Multimodal Data Should Be Graphified
  466. Exact Memory-Time Optimization for Prefix-Cached Language Model Serving
  467. A Controlled Audit of Personal AI Memory for Rating Prediction
  468. A Two-Stage Cascade for Near-Real-Time Forest Anomaly Detection from Sentinel-1 SAR Time Series
  469. Dynamic LLM Routers are Often Misguided
  470. FiberGeoText: A Vision-Language Model for Population- Level Organization of Superficial White Matter
  471. Learning Query Encoders Can Be Hard Even When Vector Retrieval Is Geometrically Easy
  472. Inner Momentum for Differentially Private Muon
  473. TPBench: A Turning-Point Benchmark for Dialogue Compression
  474. Differential Privacy of Gradient Descent on Perturbed Objectives
  475. Ego2World: Compiling Egocentric Cooking Videos into Executable Worlds for Belief-State Planning
  476. Characterizing the Performance Gap in Human Activity Recognition for Older Adults
  477. Self-Supervised Scaling of Terminal Environments for Scientific Domains
  478. MuonIO: Principled Norm-Aware Descent for Embedding Tables and Language Model Heads
  479. Label-Efficient Time Series Classification at Scale: A Dual-Stream OSSE-LSTM with Counterfactual Attribution
  480. GeoScaffold: Learning Compact Geometric Latents via Reconstruction for Efficient Vision-Language Navigation
  481. Decoupling Memory from Context: Structured Memory for Token-Efficient Test-Time Continual Learning
  482. Large language models exhibit unreliable updating of clinical judgment as patient evidence evolves
  483. RAOA: Alternating-Operator Neural Computation with Programmable Radio Propagation
  484. DataWeave: Deploying Human-LLM Analytics for Exploratory Structured Data Analysis
  485. One Photon, Many Worlds: Posteriors and Predictions with Single-Photon Cameras
  486. Asterism: Exploring and Synthesizing Scattered Observations into Literature-Grounded Hypotheses and Theories
  487. Mind the Refinement Gap: When Safe High-Level Robot Plans Produce Unsafe Executions
  488. AIGS: Adaptive Incremental Gating System for Online Representation Learning in Non-Stationary Data Streams
  489. Capturing Dynamics: The 4D Facial Expression Intensity Dataset
  490. When Normalization Selects the Sign: Auditing Robustness Ablations in Quantum Attention
  491. Batched Speech Decisions Without Decoding: Single-Token Supervision Lets a Frozen LLM Hear Beyond the Transcript
  492. Where Quantum Fourier Sampling Stops Short: A Three-Gate Audit Protocol for Delay-PUF Security Models
  493. A hybrid CNN-adjoint optimization framework for reconstruction of viscoelastic tissue properties in magnetic resonance elastography
  494. Online Verification of Language Model Responses Under Cost Constraints
  495. Designing the Future of User Feedback for Generative AI
  496. Lost in the Request: How Communication Variation Disrupts Retrieval and Action in Email Agents
  497. Quantifying the Value of Constructive Induction, Knowledge, and Noise Filtering on Inductive Learning
  498. What Is Lost in Post-Training? Default Collapse and the Loss of In-Context Steerability Across Diverse Perspectives
  499. Learning When to Commit from Partial Speech for End-to-End Simultaneous Speech Translation
  500. Seer: Maximum Likelihood Regression for Learning-Speed Curves
  501. Time Series Forecasting Benchmarks Need Scenario-Grounded Stress Testing