mlsanityData to 14 Oct 2026

Papers

New on arXiv

Sat 26 Sep 2026

558 papers

  1. Splicing Hints into Reasoning Traces: Learning from Demonstrator FailuresLLM reasoning → Reinforcement learning for reasoning
  2. GameSmith: Scaling Zero-Sum Self-Play with Synthesised GamesSynthetic data and self-improvement for LLMs → Self-play RL for reasoning
  3. Posterior Flow MatchingDiffusion models → Flow matching, time series and science
  4. PsyBer-Agent: A Psychology-Driven User Agent for Recommender SystemsLLM agents and tool use → Agentic recommender systems
  5. Kairos: Toward a Byte-Level Foundation Model for Every ModalityMixture-of-experts → Multimodal MoE models
  6. Axiom Satisfiability of Linear Rewards in AlignmentLLM alignment / RLHF / preference optimisation → Reward modeling and RLHF
  7. Certainty Is Not Just Correctness: Rethinking Token-Level Certainty in LLM ReasoningLLM reasoning → Reasoning benchmarks and evaluation
  8. Boundary-State Control for Tool-Using Language-Model Agents: Commit-Time Consistency under State DriftLLM agents and tool use → Agent threats, guardrails and governance
  9. Backdoor in the Loop: Compromising Agentic Search via Malicious RetrieversRetrieval-augmented generation → Privacy and security of RAG +2
  10. From Static Policies to Adaptive Priors in Offline Reinforcement LearningOffline RL and imitation learning from data → Offline RL algorithms
  11. Beyond Symmetric Agents: Cognitive Diversity and Multi-Agent Debate in Small Language ModelsLLM agents and tool use → Social simulation and multi-agent debate
  12. What Must a World Model Distinguish for Planning?World models → World models for language agents
  13. Residual Diffusion Implicit ModelsDiffusion models → Inverse problems and image restoration
  14. Trust and Task Completion in the World of Consumer AI AgentsLLM agents and tool use → Social simulation and multi-agent debate
  15. The Epistemics of Agent Memory: Measuring, and Governing, the Consolidation Decision in Long-Horizon LLM AgentsLLM agents and tool use → Agent memory systems
  16. Saturation-Insensitive Dueling Bandits with General Function ApproximationLLM alignment / RLHF / preference optimisation → Reward modeling and RLHF
  17. A statistical theory for blind denoising: minimax estimation of the noise level from a single sampleDiffusion models → Diffusion foundations and sampling
  18. X-Tree: Tokenizing Reusable Experience for Efficient Agent GeneralizationLLM agents and tool use → Agent skill learning and evolution
  19. Certified Long-Horizon Code Agent Evolution via Validation-Gated Skill OptimizationLLM agents and tool use → Agent skill learning and evolution
  20. ReVision3D: Attribution-Guided Recursive Self-Improvement for 3D Medical PerceptionSynthetic data and self-improvement for LLMs → Self-improving MLLMs and synthetic data
  21. Feasible Flow Matching for Graph Reconstruction via Within-Sampling Primal-Dual GuidanceDiffusion models → Flow matching, time series and science
  22. Self-Confirming Superposition Traps in Reinforcement LearningDeep reinforcement learning methods → Deep RL algorithms and representations
  23. Relic: From Multi-Agent Collaboration to Persistent Organizational CapabilityLLM agents and tool use → Software engineering coding agents
  24. The Commit-Abstain Circuit: Why Language Models Hallucinate Instead of AbstainingMechanistic interpretability → Probing deception, bias and beliefs
  25. Beyond Token Savings: A Systematic Study of Context Compression in LLM AgentsLLM agents and tool use → Software engineering coding agents +1
  26. Environmental Impact of Generative and Agentic AI: An in-Depth Analysis and Green SolutionsLLM agents and tool use → General LLM agent frameworks and benchmarks
  27. Learning Geometry-Aware Virtual Fixtures From Sparse DemonstrationsOffline RL and imitation learning from data → Robot learning from demonstration
  28. Constrained Flow Policy Updates: A Generalized Schr"odinger Bridge ViewDeep reinforcement learning methods → RL with diffusion and flow models
  29. Theory of Scene: Breaking the Symmetry Trap in Multi-Agent LLM CoordinationLLM agents and tool use → General LLM agent frameworks and benchmarks
  30. Last-Iterate Guarantees for Online Reinforcement Learning in Structured Constrained MDPsDeep reinforcement learning methods → RL theory and sample complexity
  31. Counting on Thinking: Tracing Evidence Integration in Language ModelsLLM reasoning → Reasoning benchmarks and evaluation
  32. The Geometry of Logic: Stratification Induces Semantic Structure and Robust ReasoningLLM reasoning → Mathematical reasoning
  33. Diagnosing Sampled LLM Reasoning in Formal Geometry: Coverage, Realization, and Validity EvidenceLLM reasoning → Mathematical reasoning
  34. Adaptive Latent Capacity for World ModelsWorld models → Latent predictive world models
  35. Vision-Language Agents for Active Perception in Optics LaboratoriesMultimodal LLMs / vision-language models → Embodied and 3D vision-language models +1
  36. Planner-as-Router: Joint Plan-Time Model Routing for Cost-Efficient Multi-Agent WorkflowsLLM agents and tool use → General LLM agent frameworks and benchmarks
  37. Allspark: Weak to Strong Transfer via Alternating Chain of ThoughtLLM reasoning → Reinforcement learning for reasoning
  38. Constraints Are Graphs, Not Chains: Exact Decoding for Diffusion Language ModelsDiffusion models → Discrete and text diffusion
  39. When Less Compute Is More: Adaptive Early Exit Improves Pretrained Outlier DetectionEfficient LLM inference and serving → Early exit and edge inference
  40. Optimizing H-Graph Hybridization for Diffusion-Guided RRTDiffusion models → Diffusion and flow policies for robotics
  41. StraTune: Adaptive Selection of Revision Operators for Self-Evolving LLM SkillsSynthetic data and self-improvement for LLMs → Self-rewarding alignment and self-training
  42. Multimodal LLMs Outperform Pathology Foundation Models in Cross-Domain Histological SimilarityMultimodal LLMs / vision-language models → Medical vision-language models
  43. ASCEND: Personal AI Agents for Autonomous Scientific Computing Across HPC Clusters and GPU WorkstationsLLM agents and tool use → Software engineering coding agents
  44. RoboFoundry: System-as-Policy Evolution for Self-Learning Embodied AgentsLLM agents and tool use → Agent skill learning and evolution
  45. ARSM: Auto-Regressive State Machine for Agentic Reasoning CompressionLLM reasoning → Reasoning for agents, planning and code +1
  46. Beyond Temporal Smoothing: Spatial Energy Budgets Stabilize One-Step Diffusion EditingDiffusion models → Image generation, text-to-image and editing
  47. Scanning While Imagining: A Scene-Graph World Model for Robotic Ultrasound NavigationWorld models → Medical and surgical world models
  48. Permutation-Equivariant Flow Matching for Alignment-Free Neural Weight GenerationDiffusion models → Flow matching, time series and science
  49. UniCache: Task- and Type-Aware KV Cache Compression for Unified Multimodal ModelsEfficient LLM inference and serving → KV cache optimization
  50. Retimed Bellman Flows: Escaping the Impossible Triangle of Velocity BootstrappingOffline RL and imitation learning from data → Diffusion and flow policies for offline RL
  51. Improving LLM Collaboration via Multi-Agent Preference LearningLLM agents and tool use → Reinforcement learning for LLM agents +1
  52. Routing Drift Alone Does Not Diagnose Failure in Merged MoE LLMsMixture-of-experts → Expert routing strategies
  53. Rank Collapse Is Recoverable, Growing |Q| Is Not: Out-of-Sample Early Warning for Value Divergence in High-UTD Soft Actor-CriticDeep reinforcement learning methods → RL theory and sample complexity
  54. Right Answer, Wrong Reason: Accuracy, Consistency, and Consensus Are Misleading Indicators of LLM Faithfulness in Clinical Decision SupportLLM reasoning → Clinical and biomedical applications
  55. USAI-Quant: A Quantitative Reasoning Benchmark for Vision-Language Models in Built EnvironmentsLLM reasoning → Spatial and embodied reasoning +1
  56. Overwhelmed by Choice: Studying LLM Decision Making at ScaleLLM reasoning → Test-time compute and efficient thinking
  57. Mind the Spike: Mechanisms and Brittleness of Visual Massive Activations in Large Vision-Language ModelsMultimodal LLMs / vision-language models → Visual token pruning and efficiency
  58. Decision-Sufficient State Representations: Measuring and Reducing Write-Time RegretLLM agents and tool use → Agent memory systems
  59. Nutri-ATLAS: Embodied Agent for Tabulated Lookup and Assistance for Smarter nutritionLLM agents and tool use → Medical and scientific domain agents
  60. Re-derivability Decides What a Staged Agent Pipeline Recovers After an Upstream FaultLLM agents and tool use → Software engineering coding agents
  61. PlanGuard: A Guardrail for Multi-Step Plan Safety in Embodied AgentsLLM safety and security → Prompt injection and agent security
  62. AgentHabit: Characterizing Distinct Behaviors of Agents on Everyday TasksLLM agents and tool use → Social simulation and multi-agent debate
  63. T5: Twin-Critic Training for Token-Level Thoughts in Reinforcement Mid-TrainingLLM reasoning → Reinforcement learning for reasoning
  64. CT-OPD: Counterfactual Trace On-Policy Distillation for Diffusion Vision-Language ModelsMultimodal LLMs / vision-language models → Embodied and 3D vision-language models
  65. OmniMoE-VL: A Sparse Vision-Language Model with Coupled Visual-Depth RoutingMultimodal LLMs / vision-language models → VLM benchmarks and visual reasoning +1
  66. Copper-Policy: Focus on the Representation for Robust Robot ManipulationWorld models → Embodied and robot world models
  67. On the Behavioral Traits of LLM AgentsLLM agents and tool use → Social simulation and multi-agent debate
  68. An Empirical Study on What Matters for Viewpoint-Generalizable Policies in Visual Imitation LearningOffline RL and imitation learning from data → Robot manipulation imitation learning
  69. From Feed-Forward to Flow: Unifying Reconstruction and Generation Is Easier Than You ThinkWorld models → Video-generation world models and simulators
  70. Readout is not Recovery: Dissociating Coordinate Emission from Visual-Corruption Repair in Vision-Language ModelsMultimodal LLMs / vision-language models → Visual token pruning and efficiency
  71. Adaptive Consistency Graph for Long-Horizon AgentsLLM agents and tool use → General LLM agent frameworks and benchmarks
  72. CUA-Sandbox: Efficient Environments for Computer-Use Agent Reinforcement LearningLLM agents and tool use → Reinforcement learning for LLM agents +1
  73. Retrospective Distillation Attribution via Normalized Response SimilarityDeep reinforcement learning methods → RL training of language and agent models
  74. Self-Evolving Multi-Agent Symbolic Discovery for Financial Fundamental AnalysisLLM agents and tool use → Financial trading and forecasting agents
  75. Gradient-Guided Decoupled Adaptation for Geospatial Vision-Language ModelsMultimodal LLMs / vision-language models → Embodied and 3D vision-language models
  76. Plan-to-Synthesis: Cross-City Human Mobility Generation via Semantic Latent Flow MatchingDiffusion models → Diffusion and flow policies for robotics
  77. SkillVine: Agent Skill Evolution via Branching ExplorationLLM agents and tool use → Agent skill learning and evolution
  78. Distributionally Robust Average-Reward Reinforcement Learning: Finite-Sample Guarantees under Weak CommunicationDeep reinforcement learning methods → RL theory and sample complexity
  79. Scaling Properties of Same-Family On-Policy DistillationLLM reasoning → Reinforcement learning for reasoning +1
  80. BiasReducer: Adaptive Bias Mitigation for Reward ModelsLLM alignment / RLHF / preference optimisation → Reward modeling and RLHF
  81. LLM Alignment--Utility Asymmetry under Semantic-Preserving TransformationsLLM alignment / RLHF / preference optimisation → LLM preference optimization methods
  82. Despite Instructions: Frontier Agents Improvise Covert Channels at Test TimeLLM agents and tool use → Agent threats, guardrails and governance
  83. CAIRN: Dynamic Fact-Intent DAGs for Multi-Agent ExplorationLLM reasoning → Reasoning for agents, planning and code +1
  84. SEES: A Self-Evolving Embodied System via Failure-Guided VLA Policy AdaptationVision-language-action models → Failure recovery and robustness
  85. IGSD: Environment-Verified Hindsight Self-Distillation for Search AgentsKnowledge distillation → On-policy and self-distillation for LLMs
  86. World Agent: Can Language Models Keep a World Running?World models → World models for language agents +1
  87. Silent Failures in Agentic Security Evaluation: A Validated Harness for Tool-Call Mediation Under Indirect Prompt InjectionLLM agents and tool use → Agent threats, guardrails and governance +1
  88. MM-OPD: Towards One More Bottleneck Between Perception and ReasoningLLM reasoning → Multimodal reasoning +1
  89. Self-Evolving Time-Series Forecasting Agents with Episodic Memory and Online Policy LearningLLM agents and tool use → Financial trading and forecasting agents +1
  90. RCVLA: 4D Radar-Grounded Semantic Reasoning and Trajectory Arbitration for Autonomous DrivingEnd-to-end autonomous driving → VLMs for autonomous driving
  91. The GUI Is Not the State: Diagnosing State Aliasing in GUI World ModelsWorld models → World model security and GUI agents
  92. When Better Gets Worse: Improvement Fidelity for Self-Improving Agents in Adaptive WorldsSynthetic data and self-improvement for LLMs → Self-evolving LLM agents
  93. Expected Reasoning-Step Return Unifies On-Policy Learning from Rewards and TeachersLLM reasoning → Reinforcement learning for reasoning
  94. Concepts Complement Dense Semantics: Learning Compact Sparse Spaces for Text-Image RetrievalMultimodal LLMs / vision-language models → Document understanding and multimodal RAG
  95. Learning from a Thoughtful Teacher: Adaptive On-Policy Self-Distillation for Mathematical ReasoningLLM reasoning → Prompting and chain-of-thought reasoning +1
  96. Timestep Weighting: A Hidden Key to Effective ELBO-Based Flow-Matching RLDiffusion models → Diffusion and flow policies for robotics +1
  97. Distance-KV: Exploiting Relative Distance for Efficient Long-Context InferenceEfficient LLM inference and serving → KV cache optimization
  98. InterTab: Interleaved Visual-Structure Alignment for Multi-Modal Table ReasoningLLM reasoning → Multimodal reasoning
  99. Contract Memory Compiler: Resolve, Then TraverseLLM agents and tool use → Agent memory systems
  100. World Models with Predictable Long-Horizon MarginalsWorld models → Latent predictive world models
All 558
  1. From Knowing to Abstaining: Bridging the Representation-Action Gap in Vision-Language ModelsMultimodal LLMs / vision-language models → VLM benchmarks and visual reasoning
  2. From Scene Graphs to Answers: Selective Neuro-Symbolic Reasoning for Autonomous DrivingLLM reasoning → Chain-of-thought reasoning for driving +1
  3. Business Compromise Detection with Agentic AI and LLM-driven Knowledge DiscoveryLLM agents and tool use → Agent threats, guardrails and governance
  4. Trust the Brand, Lose Control: How Identity Hijacks LLM Agent OrchestrationLLM agents and tool use → Agent threats, guardrails and governance
  5. PF-RL: Progress Field Reinforcement Learning via Goal-Conditioned Value Geometry for Vision-Language-Action ModelsVision-language-action models → Reinforcement learning and post-training +1
  6. SWE-MILE: Asynchronous Potential-Induced Milestone Credit Assignment for Long-Horizon Software Engineering AgentsLLM agents and tool use → Software engineering coding agents
  7. ExpVoyager: Direct Experience Navigation for Dynamic Agent Skill SynthesisLLM agents and tool use → Agent skill learning and evolution
  8. DraftAttention2: Fast Video Diffusion with Low-Resolution-Guided Mixed-Precision AttentionDiffusion models → Video generation
  9. World SLAM Model: Joint World Modeling for SLAM and NavigationWorld models → World models for navigation
  10. The Alignment Paradox: How Post-Training Amplifies Confident Hallucinations in Language ModelsLLM alignment / RLHF / preference optimisation → LLM preference optimization methods
  11. "You're Right, Let Me Fix It": How LLM Agents Damage Correct Work When Falsely AccusedLLM agents and tool use → Agent threats, guardrails and governance
  12. KV-Lingo: Learning KV-Cache Translators with DistillationEfficient LLM inference and serving → KV cache optimization
  13. AnchorRep: Defending LLMs Against Cross-Model Adversarial Transfer via Representation RepulsionLLM safety and security → Prompt injection and agent security
  14. CUA-SWE: When Computer-Use Agents Meet Visual Software EngineeringLLM agents and tool use → Software engineering coding agents
  15. Think Fast, Plan Selectively: Adaptive Deliberation for Efficient Data-Driven MPCWorld models → Latent dynamics models and model-based RL
  16. HERO-MoE: Historical Expert Routing with Scale-Preserving FusionMixture-of-experts → Expert routing strategies
  17. DimPO: Dimensionality Reduction for Attention using Preference OptimizationLLM alignment / RLHF / preference optimisation → LLM preference optimization methods
  18. Groupwise Agentic Grading and Advantage Redistribution for Code Agent RLLLM agents and tool use → Software engineering coding agents
  19. Trapped by Their Own Rollouts: Understanding Aggregation--Rollout Feedback in Federated On-Policy DistillationKnowledge distillation → On-policy and self-distillation for LLMs
  20. OmniSmartHome: A Multimodal Reasoning Benchmark for Smart-Home AgentsLLM reasoning → Audio-language model reasoning +1
  21. Are Vision-Language-Action Models Robust to One-Step Observation Perturbations?Vision-language-action models → Safety and adversarial attacks
  22. Shared Autoregressive Context Can Distort Relationships in Synthetic DataSynthetic data and self-improvement for LLMs → Model collapse from synthetic data
  23. Porimon: An LLM-Based Pokémon Battle Agent Enhanced by Long/Short-Term Knowledge Augmented GenerationLLM agents and tool use → Reinforcement learning for LLM agents
  24. In-Flight KV Cache with Clean Anchors for Faster Autoregressive Video DiffusionDiffusion models → Video generation +1
  25. RIPE-MambaSpike: Resolution-Independent Spiking-State-Space Interfaces for Parameter-Efficient Event-Based VisionState-space sequence models (S4, Mamba) → Vision Mamba backbones
  26. What Does a ProcGen Generalization Gap Measure? Action Rules, Residual Entropy, and the Missing Random FloorDeep reinforcement learning methods → Deep RL algorithms and representations
  27. Fail Loudly: An Auditable Runtime for Agentic Data AnalysisLLM agents and tool use → Software engineering coding agents
  28. Beyond Dyadic Memory: Interaction-Aware Multimodal Memory with Adaptive Agentic Retrieval for Multi-Party Spoken ConversationsLLM agents and tool use → Agent memory systems
  29. MemAgent: Learning to Manage Heterogeneous Memory Providers for LLM AgentsLLM agents and tool use → Agent memory systems
  30. When Users Change Their Minds: Measuring and Repairing Intent Drift in LLM AgentsLLM agents and tool use → General LLM agent frameworks and benchmarks
  31. STR: Supervised Transcoder Replacement for Reducing Steering Side EffectsMechanistic interpretability → Steering and probing for LLM safety
  32. UnStep: Training-Free Acceleration of Causal Video Diffusion with Fewer Steps Than DistillationDiffusion models → Video generation
  33. From Anomalies to Failures: Constructing Causal Error Graphs for Agentic Trace DiagnosisLLM agents and tool use → Software engineering coding agents
  34. What Do Latent Predictive Vehicle Representations Retain? Measuring State, Geometry, and Local ResponseWorld models → Latent predictive world models
  35. Learning from Others, Acting for You: Cross-User Memory Sharing for LLM AgentsLLM agents and tool use → Agent memory systems
  36. DAAF: From Failure Localization to Editable System Assets in LLM AgentsLLM agents and tool use → Software engineering coding agents
  37. Locally Sound, Globally Insufficient: The Local-Global Gap in Multi-Hop ReasoningLLM reasoning → Reasoning benchmarks and evaluation
  38. SoFT: Soft Targets for Generalizable LLM Fine-TuningKnowledge distillation → Knowledge distillation for LLMs
  39. Beyond Prompt or Skill? Attribution-Guided Optimization of Modular LLM ProgramsLLM reasoning → Reasoning for agents, planning and code
  40. RepoMAS: Solving Progressively Specified Tasks with Issue-Driven Multi-Agent SystemsLLM agents and tool use → Software engineering coding agents
  41. When Helpful Text Hurts: Option-Redirecting Bias in Vision-Language ModelsMultimodal LLMs / vision-language models → VLM benchmarks and visual reasoning
  42. When Does Dense Retrieval Need Asymmetric Geometry? A Bias-Variance Theory of Shared and Dual ProjectionsRetrieval-augmented generation → Generation and context integration
  43. Elastic Selective Spectral Hybrids for Train-Once, Export-Many Budgeted InferenceState-space sequence models (S4, Mamba) → Core SSM architectures for language
  44. From Outcomes to Strategies: Learning Strategy Utility for Mathematical ReasoningLLM reasoning → Reinforcement learning for reasoning
  45. PC-SubMax: Efficient Prompt Compression via Regularized Submodular MaximizationEfficient LLM inference and serving → Prompt and context compression
  46. On the Pitfalls of Verbalized Confidence Priors for Calibrating Large Reasoning ModelsLLM reasoning → Reinforcement learning for reasoning
  47. PULSE: Identifying Demonstration-Utility Features with Sparse AutoencodersMechanistic interpretability → Sparse autoencoder training and evaluation
  48. REMEDY: How Far Is Video Generation from Medical Education World Models?World models → Medical and surgical world models
  49. Seeing Parts, Reasoning about Worlds: Visual Inference under Partial ObservationWorld models → World models for language agents
  50. QuacamFM: Quaternion-Constrained Flow Matching for Camera Pose EstimationDiffusion models → 3D, point cloud and scene generation
  51. ForkLeft: Entropy-First Rollouts for Prefix-Aligned Autoregressive-to-Diffusion DistillationLLM reasoning → Diffusion reasoning and test-time scaling +1
  52. Masking Frequent Tokens Sharpens Direct Preference OptimizationLLM alignment / RLHF / preference optimisation → LLM preference optimization methods
  53. Rethinking Training-Inference Mismatch in LLM Reinforcement Learning: Where It Arises and How to Correct ItLLM reasoning → Reinforcement learning for reasoning
  54. From Latents to Wires: Surgical Post-Editing on Large Language ModelsLLM safety and security → Vision-language and text-to-image safety
  55. Beyond the Manifold Hypothesis: Hybrid Spectral Parameterizations for Flow MatchingDiffusion models → Flow matching, time series and science
  56. DualGuard: Dual-Mode Quality Control for Logic-Preserving Data AugmentationLLM reasoning → Test-time compute and efficient thinking
  57. Multi-Agent System Search via Active Substructure-aware Policy OptimizationLLM agents and tool use → Reinforcement learning for LLM agents
  58. Authorization Closure Graph: Minimal Repair for LLM Agents with Evolving User InstructionsLLM agents and tool use → Agent threats, guardrails and governance
  59. CityToolVQA: Tool-Augmented Visual Question Answering for 3D Spatial Cognition in Urban Low-Altitude EnvironmentsMultimodal LLMs / vision-language models → Embodied and 3D vision-language models
  60. CyberClear: A Benchmark for LLM Agent Systems on APT Attack Chain ProvenanceLLM agents and tool use → Agent threats, guardrails and governance
  61. PluginRSI: Recursive Improvement of Agent Harnesses with Reusable PluginsLLM agents and tool use → Software engineering coding agents
  62. Carnator: Fast Text-to-Video Generation with Generation-Native Compatibility-Guided Cross-Request ReuseDiffusion models → Video generation
  63. RE-0: Verified Recursive Improvement of Embodied Code-as-Policy Agents through Local On-Policy DistillationKnowledge distillation → On-policy and self-distillation for LLMs
  64. RefAdapt-DiT: Adaptive Joint Attention for Reference-Conditioned Diffusion TransformersDiffusion models → Diffusion and flow policies for robotics
  65. Shared Worlds, Private Minds: Structured Memory for Long-Form Writing as World CreationLLM agents and tool use → Agent memory systems
  66. SkillDRE: Dual-Stage Red-Team Evolution of Agent Skills via Pre-Execution and Runtime FeedbackLLM safety and security → Prompt injection and agent security
  67. Function Over Form: Distributional Orthogonalization in Mixture-of-Experts with Replica Expert MechanismMixture-of-experts → Expert routing strategies
  68. Memory as a cache: Exact context reuse and deletion by constructionEfficient LLM inference and serving → KV cache optimization
  69. Beyond Scripted Search: Sample-Efficient Reward Discovery via Agentic Black-box OptimizationLLM agents and tool use → Reinforcement learning for LLM agents
  70. DashAct: A Progressive Diagnostic Benchmark for GUI Agents in Interactive Dashboard AnalysisLLM agents and tool use → GUI and mobile agents
  71. Measurement Boundaries in LLM Financial Agent Evaluation: Fixed-Tape Execution and Multi-Defect AuditingLLM agents and tool use → Financial trading and forecasting agents
  72. AuthorityLens: Rethinking LLM-Based Agent Systems Through the Lens of AuthorityLLM reasoning → Reasoning for agents, planning and code +1
  73. An End-to-End Latent-Rollout Approach for Pushing Few-Step ImageNet-256 Generation to FID 1.11 without Fr'echet LossesKnowledge distillation → General knowledge distillation methods
  74. DiffPTS: Rethinking Diffusion ELBO for Probabilistic Time Series ForecastingDiffusion models → Diffusion foundations and sampling
  75. Black-Box Auditing of Epistemic Reliability in Multi-Agent Debate DistillationLLM agents and tool use → Agent memory systems
  76. Gradients for Interventions and Activations for Detection: Targeted Feature Learning in Language ModelsMechanistic interpretability → SAE features and their applications
  77. Fewer Tokens, More Self-Teaching: On-Policy Self-Distillation for Extreme Visual Token ReductionMultimodal LLMs / vision-language models → Visual token pruning and efficiency +2
  78. EyeVQA: Benchmarking Ophthalmic Vision-Language Models from Recognition to Spatial GroundingMultimodal LLMs / vision-language models → Medical vision-language models
  79. From Trajectories to Grounded Preferences: Process Preference Synthesis via Interaction Element Graphs for Web PRMsLLM reasoning → Reasoning for agents, planning and code +1
  80. Enabling Timely Guidance before Skill Retrieval: Retaining Helpful Warm Tips in Agent ContextLLM agents and tool use → Agent skill learning and evolution
  81. Progressive-View On-Policy Distillation for Regional-to-Global Transfer in Multimodal LLMsMultimodal LLMs / vision-language models → VLM benchmarks and visual reasoning +1
  82. Continual Data Unlearning in Diffusion Models via Transition-based RegularizationDiffusion models → Diffusion foundations and sampling +1
  83. HyperReCo: Retrieving and Connecting Evidence with Hypergraph Neural Networks for LLM Multi-hop ReasoningLLM reasoning → Knowledge- and tool-augmented reasoning +1
  84. RLHarness: Co-evolving Procedural Skills with Reinforcement Learning for Long-horizon Multimodal ReasoningLLM reasoning → Reinforcement learning for reasoning
  85. Not All Errors Matter: Decision-Relevant Prediction Error Predicts Planning QualityWorld models → World models for language agents
  86. All On-Board: Fully On-Chip Neuromorphic Q-Learning with Embedded CartPole SimulationDeep reinforcement learning methods → Spiking and neuromorphic RL
  87. Train4Merge: A Controlled Single-Teacher Study of RL vs. SFT Teachers for OPD-Based Model MergingLLM agents and tool use → Reinforcement learning for LLM agents +1
  88. Agentsensus: Consensus-Compressed Shared Memory for Multi-Agent Story WorldsLLM agents and tool use → Agent memory systems
  89. GLIDE: Generalized Layer-wise Intrinsic Distributional Evaluation for Heterogeneous LLM AgentsLLM agents and tool use → General LLM agent frameworks and benchmarks
  90. A Solvable Theory of Pre-training Data Poisoning: Regime-Dependent Scaling ExponentsLLM safety and security → Backdoor and data poisoning attacks
  91. Prefill-Free Cross-Family KV Cache Transfer for Heterogeneous Multi-Agent LLMsLLM agents and tool use → General LLM agent frameworks and benchmarks +1
  92. LAM: Efficient Lossy Agent Memory Framework With A Retrieval-Score Error BoundLLM agents and tool use → Agent memory systems
  93. Clarify the User or Verify the World? Uncertainty Routing for Proactive AgentsLLM agents and tool use → General LLM agent frameworks and benchmarks
  94. DS-VLA: A Dendritic-inspired Vision-Language-Action Model for Robust Action ControlVision-language-action models → Efficient VLA inference and adaptation
  95. Active Data Acquisition with Side Information via Discrete Diffusion PriorsDiffusion models → Diffusion foundations and sampling
  96. Federated Subspace Guided Vision-Language-Action Policy Distillation for Non-IID Multi-Robot ManipulationKnowledge distillation → Federated, robust and private distillation
  97. RoboFFT: Finetuning generative robot policy via online reinforcement learning with forward processDeep reinforcement learning methods → RL for robot locomotion and manipulation
  98. Solving Every Step Is Not Enough: Milestone Oracles Reveal a Composition Gap in LLM Math ReasoningLLM reasoning → Mathematical reasoning
  99. Skeletons in Flow: Graph Structured Flow Matching for Human Motion PredictionDiffusion models → Diffusion and flow policies for robotics
  100. OptiArena: Can LLMs Improve Executable Algorithms under Fixed Resource Budgets?LLM agents and tool use → Software engineering coding agents
  101. Toward Agentic Optical Networks: A Vision of LLM Agent-Driven Autonomous Lifecycle ManagementLLM agents and tool use → Agentic AI for wireless networks
  102. HM-ROUTER: Joint Model and Harness Routing for Agentic SystemsLLM agents and tool use → General LLM agent frameworks and benchmarks
  103. Kernel-Based Steering of CLIP with Vision-Language Model PreferencesMultimodal LLMs / vision-language models → Prompt learning and adaptation of CLIP
  104. Devol-ONE: One Autoregressive Mixture of Transformers to Unify Vision-Language-Action and Latent World ModelingWorld models → Embodied and robot world models +1
  105. ScopeIF: Improving Scope-Aware Precise Instruction-Following in Large Language Models via Graded Reward ModelingLLM alignment / RLHF / preference optimisation → Process reward models for reasoning
  106. Contamination, Prior, or Evidence? Decomposing and Training Evidence Use in Whole-Slide Vision-Language ModelsMultimodal LLMs / vision-language models → Medical vision-language models
  107. AI Harness: Certification under Proposal-Conditioned Information for Foundation-Model AgentsLLM agents and tool use → Agent threats, guardrails and governance
  108. KeyRec: Bounded Visual Memory for Streaming and Long-Video UnderstandingMultimodal LLMs / vision-language models → Video understanding with MLLMs
  109. OneFixer: High-Quality and Consistent One-Step Autoregressive 3DGS Refinement for Driving ScenesDiffusion models → Video generation +1
  110. Uncertainty-Aware Selection of Online Algorithms with Simulator EnsemblesDeep reinforcement learning methods → RL theory and sample complexity
  111. PastForward: Faster On-Device GUI Agents via Computational Experience ReuseLLM agents and tool use → GUI and mobile agents
  112. PruneForget: Joint Unlearning and Pruning of Vision ModelsLLM safety and security → Machine unlearning
  113. CausalDriveBench: Evaluating Causal Reasoning in Vision-Language-Action Models for Autonomous DrivingLLM reasoning → Chain-of-thought reasoning for driving +2
  114. Playing to Par: Reinforcement Learning for Provably Optimal Quadrilateral Block DecompositionsDeep reinforcement learning methods → RL with diffusion and flow models
  115. Escaping Alignment: A Physical Trap Model of Best-of-N JailbreakingLLM safety and security → Jailbreaking
  116. SAMBAR: Selective Anchoring via Method of Multipliers for Balanced Knowledge Acquisition and Retention in Vision-Language-Action ModelsVision-language-action models → Reinforcement learning and post-training
  117. LLM Unlearning Evaluation with TRIAGELLM safety and security → Machine unlearning
  118. Residual Streams Read, Recurrent States Remember: The Global Workspace in Mamba ModelsState-space sequence models (S4, Mamba) → Core SSM architectures for language
  119. How Vision Transformers Represent Position: A Mechanistic Comparison of Absolute and Rotary Positional Encodings
  120. Predict Before You Modify: Learning Improvement Dynamics for Self-Evolving Agents
  121. DoSA: Dots Sparse Attention
  122. An Input-Dependent Readout Phase for Resonant Sequence Models
  123. When Forgetting Looks Like Improvement: Metric Masking in Streaming Diarizer Adaptation and the Price of Rehearsal
  124. Global Evaluation of AI and NWP Precipitation Forecasts During Atmospheric River Events
  125. Generalizable Neural Downscaling of Earth System Model Wind Fields via Continuous Dynamics Modeling
  126. Fractal Cross Product: Theory, Differentiable Implementation and Application to Medical Image Analysis
  127. GeoSET: Generalist Foundation Model for SAR-to-EO Image Translation
  128. MotionMaestro: Masked Tokenization for Unified Motion Generation
  129. Learning Social Navigation from Internet Videos in the Policy State Space
  130. Authority Before Utility: Non-Compensatory Control for Persistent LLM Memory
  131. VeriWeave Govern: Evidence-Gated Deterministic Runtime Governance for Enterprise AI Agents
  132. Governing the Edge: Automating Commercial Property and Casualty Insurance Underwriting via a Hybrid Local-Cloud Multi-Agent Framework
  133. CruxBench: A Benchmark of Information Discovery
  134. BarcodeMAE+: Rethinking Masked Pretraining and Global Representations for DNA Barcode Foundation Models
  135. Hybrid Ensemble Learning for EEG-Based Epileptic Seizure Forecasting
  136. Risk-Averse Online POMDP Planning via CVaR of the Immediate Cost with Performance Guarantees
  137. More Programs or More Rolls? Separating Coverage from Specialization in LLM Harnesses
  138. The Price of Token Boundaries: Compression Certificates and Prediction
  139. Is Human-Readable Text Necessary for Effective LLM Fine-Tuning?
  140. ReLOBGen: Replayable Limit Order Book Message Generation
  141. PACT: Pairwise-Anchored Calibrated Tuning for Single-Token Typed Decisions
  142. Lukasiewicz Neural Networks Extended: Residual Architectures and Crystallization Strategies for Interpretable Rule Extraction
  143. Low-Rank Single-Index Bandits with Unknown Links: From Matrices to Tensors
  144. SRE-Marathon: A Continuous, Change-Driven Benchmark for Autonomous Site Reliability Agents
  145. Decentralized Optimization with Cross-Coupled Mixed Affine Constraints
  146. Relative Generalization Invariance of LLM Pretraining
  147. TCMQA: A 38K-Question Traditional Chinese Medicine Benchmark with a Licensed-Practitioner Reference
  148. When Pair Count Is Not the Sample Size: What All-Pairs Agent Comparisons Estimate
  149. Model-Aware Data Selection from In-and-Out Information Interplay
  150. CAPEX: Efficiently Distilling Foundation Model Behavior into Deployable Robot Policies through Experience-Adaptive Reasoning
  151. Safety-Constrained Cascade Inference for Robust Malaria Cell Classification Under Field Corruptions
  152. Certified Interface Aliases: Exact Collisions in Vision-Language Preprocessing, and When They Exist
  153. TriDrive: Joint Driver, Vehicle, and Road Modeling for Forecasting and Driver Monitoring
  154. Oracle Gaps in Reliability Coverage: Sampling Noise or Policy Specialization?
  155. What Should Data Teach? Moving Bottlenecks Across Circuit, Store, and Use
  156. Maximum Likelihood Estimation for Entity Ranking under Iterative Synthetic Data Augmentation
  157. Adaptive Ensemble Selection for Noisy Labels on Tabular Data
  158. Distributed Hydrological Modeling in the Feature Space
  159. Clipped or Unclipped? Finite-Sample Trade-offs for Averaged SGD under Heavy-Tailed Noise
  160. DynamicDx: Evaluating Evidence Acquisition in Video-Based Diagnosis
  161. Efficient Message Passing for Partial Differential Equation Priors
  162. Optimal Nonparametric Dynamic Pricing with Censored Demand and Adversarial Inventory
  163. Synthetic Thermal Image Generation for Real-Time Animal Detection Under Low-Visibility Conditions
  164. The Key Handoff: Retrieval in Hybrid Language Models
  165. Generative Priors Conditioned on Natural Language for Bayesian Inversion in PDEs
  166. TwinS-GCN: Spectral conjugate for Spectral Graph Convolutional Networks
  167. The Impact of Stochasticity on the Rashomon Effect in Machine Learning
  168. EEG-Based Motor Imagery BCI Algorithms and Technologies: A Review
  169. Efficient Dynamic Algorithms for Graph Neural Networks with Non-Linear Propagation
  170. Phenomenon-Graph JEPA: Label-Efficient Representation Learning for Contactless Cardiorespiratory Sensing
  171. Identification of Nonlinear and Dependent Latent Factor Structure through Clique Search
  172. TRACE: Learning to Self-Calibrate Wireless Digital Twins from ISAC Measurements
  173. Precision As You Need: Stochastic Computing Is a Dense Adaptive Quantizer
  174. AgentTell: Behavioural Side-Channel Leakage in Browser-Use Agents
  175. Progression- vs Automata-based Anticipatory Monitoring of LTL over Finite Traces (Extended Version)
  176. Predicting the Next State Is Not Enough: JEPA Representations for Lean Theorem Proving
  177. Logical subspace in LLMs
  178. Linger and Lose: Knowledge Collapse in Low-Bit Language Models
  179. Rethinking the Fully Hyperbolic Vision Transformer in Polar Coordinates
  180. A Function-Level Vulnerability Score Measures Flag Rate More Than the Model: Protocol Effects on Paired Benchmarks
  181. Measurement-Gated Provenance Attenuation for Frozen EEG Representations
  182. Refreshing Less, Selecting Better: Reusing Stale Gradient Features for Efficient Influence-Based Data Selection
  183. Neural Network-Assisted Refinement of Traditional Schemes for One-Dimensional Scalar Conservation Laws
  184. Can LLMs Predict the Future? A Brier Score Analysis of Prediction Markets
  185. Improving Video Sparse Attention with Fine-grained Router and Sparse Rebasing
  186. Machine learning for the LHC physics program: a 2025-2026 stocktake
  187. Counterfactual Self-Evolving Agents for Evidence-Grounded Reasoning
  188. Are You Sure You're Sure? Two Confounds in a Sycophancy Benchmark
  189. Staying on the Attractor: Supervising Neural Surrogates of 3D Turbulence Where They Leave It
  190. SV2V-RSim: A Comprehensive Benchmark for Self-Selective V2V Cooperative Perception with Near-Realistic Data
  191. Muon Under Gradient Noise and the Limits of Orthogonalization Near Optima
  192. Is H&E Image-to-Spatial Transcriptomics Simpler Than It Looks?
  193. PolyTopoBench: A Benchmark for Complex Vector Polygon Generation from Remote Sensing Imagery
  194. FINE: Future-Informed Navigation Encoding for Data-Efficient Vision-Language Navigation
  195. Logic Gate Networks and Lookup Table Networks as Lightweight Hardware Classifiers for Inter-patient ECG Arrhythmia Classification
  196. Transfer Learning for Edge Classification on Dynamic Text-Attributed Graphs
  197. Sliced Orlicz-Wasserstein
  198. SynCo: Learning Cross-Modal Synergy by Contrasting Interaction Residuals
  199. Mask2Restore: Self-Supervised Ultrasound Despeckling via Inpainting
  200. VCRE-Fib: View-Conditioned Regional Evidence for Fine-Grained Ultrasound Grading of Schistosoma japonicum-Associated Liver Fibrosis
  201. Mend the Measurement Gap: Latent User Preference Modeling for Short-Form Video Recommendation
  202. HamiFormer: Dual-Expert Diffusion Fields with Affine Symplectic Maps
  203. Radiomap Blind Prediction under Incomplete Observation: Error Characterization and Correctable Propagation-Prior Learning
  204. FinancialAuditBench: Benchmark Construction under Differential Privacy Using Real-World Priors
  205. Over-the-Air Federated Learning in Heterogeneous Mobile Wireless Networks
  206. The Decomposition Tax: LLM Pipelines Lose Up to 40 Accuracy Points at Their Own Interfaces
  207. Unlocking Geodesic Gromov-Wasserstein Distances for 3D Modeling
  208. Learning Shuffle Ideals with Membership Queries and Contrastive Examples
  209. When Can First-Order Models of Fine-Tuning Bound Forgetting?
  210. Change the Product, Keep the Parameters: Associative Algebra Layers for Transformers
  211. Progressive Risk Estimation for Accident Anticipation
  212. OpenTumorBoard: A Real-World Benchmark of Multidisciplinary Tumor Board Discussion Trajectories
  213. Beyond Accuracy: Counterfactual Fragility and Demographic Bias in Clinical Evaluation of LLMs
  214. Finding Emotions Where They Belong: Rethinking Audio Emotion Recognition through Masked Temporal Affective Grounding
  215. Derivative-Informed Training of Neural Operators On-the-Fly via Sketched Tangent Consistency
  216. Latent Space Is Not Flat: Rethinking Latent Structure for 3D Medical Image Synthesis
  217. Understanding and Exploiting Anisotropy in Post-Training
  218. FedHV: Low-Overhead Hypervolume Weighting for Federated Multi-Objective Optimization
  219. Hierarchical Frequency-Domain Compression of Implicit Geometric Representations for Large-Scale Point Clouds
  220. Getting Motif-ated: Controllable AI Compositions from Injected Motif Prompts
  221. CLAIRE: A Schema-Grounded Hybrid Workflow for Healthcare Administrative Form Completion
  222. Learning When to Recur: Token-Adaptive Recursion for Imbalanced Ophthalmic Domain Incremental Learning
  223. CollisionGAT: Controller-Agnostic One-Step Collision Screening for Multi-Agent Motion
  224. Learning response-aware patient dynamics for respiratory support
  225. Agentic Network Traffic Monitoring
  226. Robust Bayesian Optimization with Q-Exponential Surrogates
  227. Beyond Gaussian Assumptions: Distribution-Aware Channel Capacity for Effective Connectivity
  228. Forecasting Intraday USD/CAD Exchange Rate with News-Derived Monetary-Policy Signals
  229. Learnable Randomization as Commitment Against Adaptive Optimizers
  230. Continual Learning via Self-Probe Gradients
  231. C-HAT-Bench: Benchmarking Chinese AI-Text Detection Beyond Fully Generated Text
  232. Structuring Relations Among Learning Paradigms via Protocol--Objective--Resource Reductions
  233. AnchorMixGAN: Anchor-Aligned Generative Semi-Supervision for DDoS Detection in Cloud-Integrated IoT Networks
  234. Mandela-Bench: Multimodal Models Remember Canonical Images Instead of Seeing Them
  235. The Extender: A Log-Structured Transformer
  236. How Far Do Persona Effects Generalize in Language Models?
  237. Reuse or Relearn? A Spectral View of Earth Observation Foundation Models
  238. SAGE: Semantic Audio Generative Encoder
  239. Action Shaping: Policies Absorb What They Can Express
  240. Benchmarking EEG Foundation Models at Scale: Lessons from 20,000 Evaluations
  241. LoCoVSR: Local Context Diffusion Posterior Sampling for Video Super-Resolution
  242. Predicting the Financial Impact of Supply Chain Risk for Major AI-Related Semiconductor Firms: A Heterogeneous Graph Patch Transformer Approach
  243. AnesTRACE: Benchmarking Intraoperative Anesthesia from Multimodal Perception to Multi-step Decision-Making
  244. REALIS: A Curated Dataset for Studying the Challenges of AI Image Detection
  245. Domain Adaptation with Target Information via Doubly-Anchored Distributionally Robust Optimization
  246. Revisiting AdaGrad in Stochastic Convex Optimization: Last Iterates, High Probability, and Lower Bounds
  247. Region-Local Copula Evidence Fusion for Heterogeneous Remote Sensing Change Detection
  248. MassAlloc Attention: Let Attention Allocate Its Own Compute
  249. ProDyGS: Dynamic Gaussian Splatting from a Single Static Monocular Camera
  250. One-Step Generative Modeling via Unbalanced Optimal Transport
  251. Learning to Refer: Client-Resolved Generation for Privacy-Aware Language Models
  252. DPAMixerSR: An Efficient Degradation-Pattern-Aware Model for Image Super-Resolution
  253. CoWindow Attention: Full Causal Coverage Is a Collective Property
  254. Flat-Consensus Diffusion for Robust Data Reshaping under Noisy Evaluator
  255. Affine Geometry of Gaussian ReLU Networks via Conditional Kac-Rice Formulas
  256. Dude, Where's My State? Execution Information Requirements for Stateful Agents
  257. PINNMorph: Evolving Online Adaptation Policies for Physics-Informed Neural Networks
  258. Focusing Condition: Inference-Time Self-Contrastive Steering Elicits Better Conditional Text Embeddings in LLMs
  259. SIFT: Enhancing Time Series Foundation Models via Semantic Invariance and Structural Fidelity Fine-Tuning
  260. STAMP: Predicting Out-of-Distribution Generalization without Target Data
  261. What Would Falsify It? A Variable Specific Evidence Standard for Mechanistic Claims About Self Explanation
  262. Refinement Symmetry in Multimodal Transformers
  263. AsynCodeBench: Benchmarking Collaboration of Asynchronous Multi-Agent Systems in Software Engineering
  264. Equivariant Neural Primal-Dual Assignment for Maximum Common Edge Subgraphs
  265. Quantization-Aware Pre-Training with Constrained Empirical Weight Distribution
  266. MixBench-TS: A Multivariate Time Series Forecasting Benchmark Where Channel Mixing Pays Off
  267. Prediction Limits and Koopman Closure of Geometry-Induced Soft State Abstractions
  268. Ask Without Telling: Local SLMs Consult Cloud LLMs Without Revealing Task Intent
  269. Extremely Fast and Compact Binary Graph Representations via Randomized Operator Sketching
  270. Schur-Neural KF: Learned Schur-Consistent Corrections to the Extended Kalman Filter
  271. Can Open-Weight Large Language Models (LLMs) Simulate Human Survey Populations? A Cross-Instrument Calibration Study
  272. Learning What to Evaluate: Correlation-Aware Decoupling for Multiobjective Bayesian Optimization
  273. MixDetect: Word-Level Localization and Quantification of AI Editing
  274. Does CoT-Pass@k Really Check the CoT? A Multilingual Mathematical Audit
  275. Intuition vectors
  276. Stabilizing the Dynamic Low-Rank Training
  277. Learning the Graph and the Embedding Together: Classifier-Independent Rewiring for Heterophilic Node Classification
  278. Levy-Driven Correspondence Estimation for Registration
  279. Not Every Term Adds New Structure: Sobolev Novelty for Symbolic Regression
  280. GAUGE: Group-Wise View-Inconsistency Rectification for Feed-Forward 4D Tracking
  281. RECAST: Recasting Vision-Language Semantics into an Actionable Cost Map for Robot Navigation
  282. MA-FPPO: Multi-Agent Flow-Pretrained Policy Optimization
  283. Age of Learning: Temporal Persistence of Prediction Errors as a Learning Signal
  284. SPACE: Sparse Predictive Attractor via Counterfactual Eviction for Streaming Video Memory
  285. Retrieved but Not Delivered: Multimodal Memory Delivery for Long-Term Agents
  286. EMIR2: Evolution-Aware Memory with Intent-Guided Multi-Round Retrieval
  287. CUE-Mem: Benchmarking Long-Term User Memory via Implicit Cues in Multimodal Conversations
  288. CAESAR: Clustering via Autonomous Embedding-Space Agglomerative Reorganization
  289. Attribution Gaps in Zero-Training LLM+OVOD Pipelines: A Fine-Grained Analysis of the CAAP--SNAP Discrepancy
  290. Compositional Objectives: Learning Structure in Structure
  291. Reading Is Not Leaking: Local, Auditable Measurement and Reduction of Inference Exposure from Public Footprints
  292. ProTTT: Learning to Learn Semantic User Memory with Test-Time Training
  293. Artificial intelligences and human scientists exhibit complementary strengths in theory building
  294. How to Reduce Whisper Hallucination
  295. CFCH: Coarse-Fine Collaborative Hierarchical Learning for Anterior Segment Disease Analysis
  296. Language as an Independent Information Layer: A Conceptual Model of Communication, Cognition and Decision-Making
  297. Feature Space Guidance for Breast Cancer Classification in DCE-MRI
  298. Harnessing Coupled Stream Completion For Human-Object Interaction Modeling
  299. Prioritizing Repeated LLM Evaluation for Hidden Failure Discovery
  300. Interpretable Physics Informed WiFi Indoor Localization: Learning an Effective Access Point Geometry and Using It to Prune
  301. Do Audio LLMs Listen Before They Act? Diagnosing Acoustic-Context Gating in Voice Agents
  302. DepthBench: Measuring How Residual Connections Enable More Computational Depth
  303. LLMAdBench: A Human Preference Benchmark for Advertising in LLM Responses
  304. Activation Flow: Manufacturing Activations for Steering
  305. SetOPD: From Few Visual Exemplars to Multimodal Candidate Sets for Remote-Sensing Open-Prompt Detection
  306. AmbiModBench: Benchmarking Gene Perturbation Prediction Beyond Shared Responses
  307. Does Transolver really need a Transformer?
  308. LocalProp: Neuro-Localized Memory-Efficient Backpropagation
  309. REFINE: A Resilient Evolution Framework for Intelligent Enterprise Alert Triage in Security Operations Centers
  310. Cross-Domain Few-Shot Writer Adaptation for Real-World Handwritten Mathematical Expression Recognition
  311. GeoCR: Learning a Generalist Cloud Removal Prior from Heterogeneous Observations
  312. Fast Differentiable SVD on GPU via Polar Decomposition
  313. Fisher Simplicity in Kolmogorov-Arnold Networks and Multilayer Perceptrons
  314. TreeRef-BFN: Equivariance-Free De Novo Molecule Generation based on 2D Topology and Internal 3D Geometry
  315. Rondo: Unsupervised Discovery of Recurring Temporal Structure
  316. Learning an Anchored Prompt Space for Continual Adaptation of Large Language Models
  317. Hearsay: Can an Auditor Trust the Record a Deployed Agent Harness Writes?
  318. Explaining Textual Entailment with Lexical Entailments: Using LLMs to Supply Lexical Relations for Formal Proofs
  319. Neural Dynamics as the Composition of Quantized Units
  320. What Should Federated LoRA Share? FedSAIL via Input-aware Subspace Alignment
  321. Towards Scalable Data Diversification for Language Model Pretraining via Leverage Score Sampling
  322. Separating Diagnosis from Disease Representation: Dual-View EEG Learning with Neural-Dynamics-Guided Deformation
  323. JEPA Learns What the Mask Leaves Unrecoverable
  324. Fast and Precise Learned Charged-Particle Trajectory Regression at the Large Hadron Collider
  325. Agnostic Smoothed Online Regression with Adversarial Responses
  326. Back-Tracking from Clarity: Self-Learning to See Text from Afar
  327. VPEvolve: A Self-Evolving Virtual Process Engineer for Computational Lithography
  328. AdaTutoRank: Learning to Rerank Document Sets via Adaptive Tutoring Optimization for RAG and Deep Research
  329. Bison: Cross-Dataset Learning for Unseen-Compound Perturbation Prediction
  330. PolyStepOR: Learning to Decide Without Optimal Decisions
  331. Adapting Nonstationary Multi-output Gaussian Processes to Bayesian Optimization
  332. Can Motion-Language Models Ground Structure? STRIDE for Evaluating the Evaluators
  333. Streamlined Reflective Evolution for Task-Adaptive Self-Refinement Pipelines
  334. Write Back the Delta: Revisiting the Same Tokens with Fresh Representations
  335. De-biasing Skeleton-based Action Recognition with Convex Hull Adaptive Shift
  336. DRAM: Delta-rule Recurrent Associative Memory for Robot Manipulation Policies
  337. What Should the Reflector See? An Empirical Study of Evidence in Reflective Prompt Optimization
  338. Self-Reports Do Not Identify Self-Models: An Identifiability Test for Counterfactual Reports
  339. Length-Independent State Tracking Under a Parallel Scan
  340. Controllable GNN Explanations via Multi-Metric Preference Selection
  341. PrismQuant: Optimal Null-Space Rotations for Grouped Quantizers
  342. HoTS: Homophily-Aware Temperature Scaling for Graph Neural Network Calibration
  343. MergeHEIR: Mitigating Multimodal Hallucinations as the Tax of Model Merging
  344. AECSF: Adaptive Ensemble Conditional Score Filtering for High-Dimensional Nonlinear Data Assimilation
  345. Recovery-Directed Symbolic Distillation of Neural Likelihoods
  346. Automatic Speech Recognition for the Basa`a Language: A Low-Resource Approach
  347. Opening LLM Judges: Recovering Preference Signals Beyond the Final Verdict
  348. Toward On-Chip Training of Spiking Neural Networks for Dense Event-Based Vision
  349. Endo-TSR: Temporal Spectral Modeling of Appearance and Motion for Endoscopic Reconstruction
  350. SinBrief: A Hybrid Framework for Abstractive Text Summarisation of Sinhala Legal Documents
  351. FA-Bench: A Benchmark for Phone- and Word-Level Timestamp Accuracy in Forced Alignment and ASR on Clean and Noisy Speech
  352. Convergent Plug-and-Play Image Restoration with Annealed Noise Levels
  353. SCLATE: a Substrate for Continual-Learning Agent Training and Evaluation
  354. Reward Hacking and Agent Containment Failure: A Monte Carlo Study Based on the 2026 Hugging Face Incident
  355. RefCompose: Multi-Reference Image Generation via LoRA-Conditioned Diffusion
  356. Using Machine Learning to Investigate Predictors of Fasting Blood Glucose: Insights into Circadian Timing and Age Interactions
  357. TimeES: Probabilistic and Deterministic Time Series Forecasting via Evolutionary Spectra
  358. PlurVA-LLM-2026 Shared Task Track-1: Pluralistic Value Alignment in LLMs via Multilingual Fine-Tuning and Threshold Calibration
  359. Traceprop: Training Data Attribution in a Single Pass
  360. STRIDE: State-Transition Representation via Increment Dynamics and Evolution
  361. Graph Memory: Spectral Associative Memory via Dirichlet Energy
  362. StegGNN: Learning Graphical Representation for Image Steganography
  363. Editable Map-Conditioned Trajectory Generation for Human Mobility Simulation
  364. Analog-Friendly Predictive Coding without Activation Derivatives
  365. ALLOT: Budgeted Hybrid-Memory Routing for Knowledge Updates in LLMs
  366. OpenMASC: An Open-Source Pipeline for Cross-Trajectory Metal-Aware Sampling and Correction in Accelerated MRI
  367. A Comparative Analysis of Attention versus State-Space Models for In-Context Learning
  368. SGA-Flow-GRPO: Spatial Gradient-Guided Credit Assignment for Flow-GRPO
  369. When the Merge Coefficient Stops Mattering: Proximity Regularized Merging for Continual LoRA Adaptation
  370. Language Distances are Practical for Equitable Cross-Lingual Transfer
  371. Active Feature Acquisition With Incomplete Training Data
  372. FoundDSR: A Generalizable Foundation Model with Guided 2D Gaussian Splatting for Depth Super-Resolution
  373. Superposed Inference for Hyperdimensional Computing
  374. Phase Space Attention:A Hairer Lift Circumvents the Single-Layer Induction Obstruction
  375. What Can a Leaderboard Certify? Compositional Controllability for Fair Evaluation and Training of Biomedical Literature-Review Agents
  376. One Perception, All Maneuvers: Directional Traffic Signal Understanding for Maneuver-Level Signal Intent Prediction
  377. Delayed Supervision for Test-Time Language Models
  378. PhiFold: Towards Dynamic Protein Design with Physics-Structured Covariance Modeling
  379. On the Capability and Limitation of Hard Prompt
  380. When Does Synergy Help Active Feature Acquisition? A PID-Based Study
  381. Refresh or Realize? Compute Allocation in Drifting Models
  382. FUND: Density Flow for Sampling Unnormalised Distributions
  383. TRAP: Understanding and Mitigating Privacy Memorization in Language Models
  384. Affordance-Conditioned Decision Making: Bridging the Semantic-Spatial Gap in Zero-Shot Cross-Floor Vision-and-Language Navigation
  385. A Journey to the Edge of Stability
  386. Neural ODEs Meet Concurrent Learning: Stable Online Learning with Lyapunov Guarantees
  387. Supporting and Performing Culture from the Inside
  388. HeroFrame-Bench: Reference-Anchored Evaluation via Rubric--Ranking Co-Evolution for Movie Hero Frame Selection
  389. SIMANF: Sample Free Learning of Unnormalized Distributions via Simulated Annealing in Normalizing Flows
  390. HyperLabel: Multi-Label Classification via Hypergraph-Based Label Correlation Modeling
  391. Topology-Adaptive Hyperbolic Graph Attention Networks Guided by the Hyperbolic Sombor Index
  392. When Does a Skill Add Value? Task-Conditional Gain Prediction for Selective Skill Use
  393. FSS-UBrain: Multi-region Few-Shot Brain Tumor MRI Segmentation
  394. Learning Through Game: Skewed Transfer of Tabular Knowledge to Strengthen Image Model
  395. Certification Frontiers for Gaussian LoRA: Independent Priors, Posterior Risk, and Prediction-Preserving Balancing
  396. Overfitting of Spectral Gradient Descent: How Matrix Geometry shapes Generalization and Implicit Bias
  397. Before Answering: Evidence Sufficiency under Size-Matched Memory Construction
  398. Self-Reconstruction Dynamics for Autoencoder Reconstruction Refinement
  399. When Can Old Evaluations Certify a New Model? Label-Efficient Release Decisions under Evaluator Drift
  400. LANTERN: Illuminating Hidden Mathematical Knowledge in Language Models
  401. Representation Editing for Multimodal Test-Time Adaptation
  402. Why Directly Learning Periodic Trajectories Can Fail
  403. RoboSTAR: Next-Scale Autoregressive Sign Language Translation for Humanoid Robots
  404. Anytime-Valid LLM Leaderboards via Benchmark-weighted and Block-Factorized e-Processes
  405. Certifying Interventional Agreement Among Observationally Equivalent Causal Models
  406. AutoPDEBench: Benchmarking LLM Auto-Research for Neural PDE Solver Design
  407. Two-Stage Multi-View Gait Recognition with a Re-Embedding Network
  408. Residual Transferability in Neural Image Watermarking
  409. Arithmetic Simplicity in Stochastic Gradient Methods
  410. KinyaMed: Seeds, Not Rows -- What a Corpus Requirement Written in the Wrong Unit Fails to Constrain
  411. "Where Can I Trust You?": Boundary-Aware Evaluation of Surrogate Fidelity
  412. CompassPlay: Rewarding the Proposer for Where It Moves the Solver
  413. LaMET-Agent: An Agent Framework for Large-Momentum Effective Theory Analysis
  414. RAO-Nav: Probing Omni-Language Models for Zero-shot Semantic Audio-Visual Navigation
  415. Geometry-Preserving Blind Watermarking for Raw 3D Point Clouds
  416. A bilingual AI audiologist built through rubric-guided closed-loop engineering outperforms human audiologists in a blinded evaluation of simulated cases
  417. A model of rational interlocutors: Unification of comprehension and production
  418. DP-Rec: Towards Dynamic Patching for Efficient Long-Sequence Recommendation
  419. Rank Confidence Sequences:Anytime-valid Leaderboards
  420. Fracast-0: Fractal Weight Sharing for a Time Series Foundation Model with Only 85K Parameters
  421. Witness: Discovery, Deciphering, and Epiphany in Interactive Puzzle Environments
  422. DegreeSpar: Structured Degree Sparsity for Efficient Secure Transformer Inference
  423. Instruct, Not Answer: Using Instruction Privileges in On-Policy Context Distillation
  424. Generalization and Memorization along the Learning Trajectory of Neural Language Models: A Geometric Account of Categorization
  425. Agents as Software: A Programming Languages Agenda for Agent Reliability
  426. SparSP: Exploiting Communication Sparsity for Sequence-Parallel Video DiTs
  427. The Judge Is Not Its Twin: Post-training makes a model's writing more predictable but barely moves its taste, as a judge, toward predictable writing
  428. High-Probability Guarantees for SGD under β-Heavy-Tailed Gradient Noise
  429. CoMemBench: Benchmarking Collaborative Memory Boundaries across Multi-Agent Workflow Topologies
  430. Evaluating Single and Multi-Omics Based Explainable Artificial Intelligence (MOXAI) for Molecular Subclass Classification of Adult-Type Diffuse Gliomas
  431. Presence Is Not Faithfulness: Figurative Vehicle Intrusion in Text-to-Image Generation
  432. Rethinking Cross-Channel Importance in Time-Series Forecasting
  433. Scalable In-Domain Self-Supervised Foundation Model for Dense Representation Transfer in High-Resolution Plant Imaging
  434. Binaural Audio-Visual Instance Segmentation
  435. Analytic-Walk Rotary Positional Encodings for Graphs
  436. Federated 3D Gaussian Splatting for Large-Scale Scene Reconstruction at Wireless Edge
  437. Efficient Support Recovery of Mixtures of Sparse Linear Classifiers with Fewer Measurements
  438. Noisy Test-Time Reinforcement Learning for Code LLMs
  439. Uncertainty Quantification of Next Generation Reservoir Computing with Applications to Memory-Driven Dynamical Systems
  440. CAFE: Counterfactual Prediction via Fast Posterior Estimation
  441. FOCUS: Fixed-Confidence Online Causal Learning Using Sequential Adaptive Interventions
  442. PQR3D: Progressive Query Refinement over Reference-Conditioned Temporal Windows for Multi-View 3D Object Detection
  443. Typed Decision Models: An Early Evidence Audit and Evaluation Checklist
  444. AquaBEV-Nav: Learned BEV Occupancy for Underwater Navigation and Exploration
  445. A Unified Optimism-Agnostic Framework for Linear Bandits over Spherical Action Sets
  446. Spectral Reversal: Counteracting Singular Value Bias for Graph Prompting
  447. REALM: Regime-Switching, Explainable, and Activation-Induced Linear Models
  448. Checking Leakage Witnesses versus Certifying Bounded Non-Leakage
  449. Latency-Aware Client Assignment for Parallel Split Learning With Global Sampling
  450. Gaussian Image Steganography via Parameter-Domain Keyed Embeddings
  451. What Should We Freeze? Guarded Freezing: Connectivity Shapes the Fine-Tuning of Pretrained Models
  452. READ-Bench: Benchmarking Historical Instance Retrieval for Time-Series Diagnosis
  453. Using LMs to Model the Effects of Context and Coreference during Sentence Comprehension
  454. Handwritten Digit Leakage from Smartphone Motion Sensors Across Unseen Users and Phone Models
  455. Empowering Hybrid Attention Models on NPUs
  456. How Reusable Are Benchmarks with Richer Feedback?
  457. An Attention-Driven Heterogeneous GNN Model for Credit Card Fraud Detection
  458. Emergent One-Third Scaling Law as Attention Tries to Concentrate