mlsanityData to 7 Oct 2026

Papers

New on arXiv

Wed 7 Oct 2026

669 papers

  1. Why LLM Agents Favor Their Group: Stakes, Observed Norms, and ReputationLLM agents and tool use → Social simulation and multi-agent debate
  2. Curating Always-Loaded Context for LLM Agents: A Capacitated Assortment Model with Censored FeedbackLLM agents and tool use → Agent memory systems
  3. How Narrative Wrapping Affects LLM Refusal: A Cross-Language Benchmark and DefenseLLM alignment / RLHF / preference optimisation → Safety alignment and jailbreak attacks +1
  4. Invariant-Measure Reasoners: Stable Representations for Latent ReasoningLLM reasoning → Test-time compute and efficient thinking
  5. Back in Style: A Sociolinguistic Approach to Authoring and Measuring Persona Fidelity in User SimulationLLM agents and tool use → Social simulation and multi-agent debate
  6. iAm.md: Robot Skill Self-Assessment through Agentic Introspection for Unknown Open-Vocabulary DomainsLLM agents and tool use → Multimodal and embodied agents
  7. Cross-Provider Review as a Runtime Contract for Coding Agents: A Controlled Pilot and Fault-Injection StudyLLM agents and tool use → Software engineering coding agents
  8. RH-Detect: A Unified Benchmark for Reward Hacking DetectionLLM alignment / RLHF / preference optimisation → Reward hacking and misspecification
  9. StoreBench: A Live-Commerce Environment for Evaluating and Training Autonomous Operator AgentsLLM agents and tool use → Reinforcement learning for LLM agents
  10. World-Model Policy Arbiter for Goal-Conditioned Reinforcement LearningWorld models → Latent dynamics models and model-based RL +1
  11. Speedbumps: Rejection Attacks on Speculative DecodingEfficient LLM inference and serving → Speculative decoding
  12. Power Side-Channel Membership Inference Attack on Embedded Machine LearningLLM safety and security → Membership inference attacks
  13. Less from More: Reinforcing Sparse Video Reasoning from Dense ReferencesLLM reasoning → Multimodal reasoning
  14. SPLIT-RL: Staged Perception-Language Reasoning Training with Claim-Level AdvantagesLLM reasoning → Multimodal reasoning
  15. Stochastic Teacher Intervention for Agentic On-Policy DistillationLLM reasoning → Reinforcement learning for reasoning +1
  16. Enabling Preference-driven Unlearning in Few-step Distilled Text-to-Image Diffusion ModelsDiffusion models → Diffusion foundations and sampling +2
  17. RFChipAgent: Multi-Agentic AI Flow for Analog/RF Chip DesignLLM agents and tool use → Agentic AI for wireless networks
  18. Cross-Embodiment Robot Foundation World Models with Latent ActionsWorld models → Embodied and robot world models
  19. Velocity Scaling in Flow MatchingDiffusion models → Flow matching, time series and science
  20. Skill-SLM: Agent Skill-driven Small Language Models for Reliable Robot OperationLLM agents and tool use → Agent skill learning and evolution
  21. NavGPT-3: Harnessing Context in a Hierarchical Navigation RuntimeLLM agents and tool use → Multimodal and embodied agents
  22. Plan-and-Patch: Diffusion Language Models for Agentic PlanningDiffusion models → Diffusion and flow policies for robotics +1
  23. VICO: Visual Environments Co-Evolving for Vision-Language Model ReasoningLLM reasoning → Multimodal reasoning +1
  24. MemoWM: How World Models Change What Agents Need to RememberWorld models → World models for language agents
  25. BRANCH: Bypassing Multi-Scanner AI GuardrailsLLM reasoning → Reasoning for agents, planning and code +1
  26. Nullify: Null-Space Activation Steering for Training-Free LLM UnlearningLLM safety and security → Machine unlearning
  27. Coverage-Aware Reasoning with Medical Tokens for Diagnosis PredictionLLM reasoning → Clinical and biomedical applications
  28. How Many Directions Must a Truncated Diffusion Sampler Retain? Matching Bounds Under Power-Law SpectraDiffusion models → Diffusion foundations and sampling
  29. Visible Reasoning Is Not a Universal Optimizer: Persona- and Thinking-Dependent Effects in Analytics Code GenerationLLM reasoning → Prompting and chain-of-thought reasoning
  30. Speaking the Navigator's Language: Trajectory-Grounded Instruction Translation for Frozen Aerial VLN AgentsLLM agents and tool use → Multimodal and embodied agents
  31. The Harness as the Only Mutable Surface: Compliance-Bounded Self-Evolution of LLM Agents in Credit Pipelines, with a Measured Admission GateLLM agents and tool use → Software engineering coding agents +1
  32. Agent4RE: A Self-Refining Multi-agent Framework for End-to-End Software Requirements Engineering and BenchmarkingLLM agents and tool use → Software engineering coding agents
  33. Safe at One Loop, Risky at Another: Aligning Safety Across Recurrent Depths in Looped Language ModelsLLM alignment / RLHF / preference optimisation → Safety alignment and jailbreak attacks +1
  34. Recurrent Self-Improvement: Dynamic Cross-Loop On-Policy Distillation for Looped Language ModelsSynthetic data and self-improvement for LLMs → Self-rewarding alignment and self-training
  35. TestJack: Should you trust the results in coding benchmarks? Agentic Coding Benchmarks Auditing via Evaluator EvolutionLLM agents and tool use → Software engineering coding agents
  36. When Routing Reveals Membership: Privacy Leakage from MoE Router TelemetryMixture-of-experts → Expert routing strategies +1
  37. Verification and Self-Improvement in Agentic AI: Foundations and LimitsLLM agents and tool use → General LLM agent frameworks and benchmarks
  38. Code Understanding is a Bottleneck for Coding AgentsLLM agents and tool use → Software engineering coding agents
  39. From Investigation Failures to Reliable SOC Agents: Understanding and Improving LLM-Based Alert TriageLLM agents and tool use → Agent threats, guardrails and governance
  40. Beyond Type-checking: Towards Holistic Evaluation of Formal Specification GenerationLLM agents and tool use → Software engineering coding agents
  41. RoboPrompt: Intuitive Robot Policy Steering with Sparse Human InputDiffusion models → Diffusion and flow policies for robotics +1
  42. Long-WAM: Scaling the Context of World-Action ModelsWorld models → Embodied and robot world models
  43. Rephrase Before You Act: Characterizing and Mitigating Language Sensitivity in Vision-Language-Action ModelsMultimodal LLMs / vision-language models → Prompt learning and adaptation of CLIP
  44. GRACE: Generation-aware latent compression for efficient video generationDiffusion models → Video generation
  45. Why Forget-Only Unlearning Needs MemorizationLLM safety and security → Machine unlearning
  46. RoboJEPA: Scaling Robotic Latent World ModelsWorld models → Embodied and robot world models
  47. Agentic RSR: Real-to-Sim-to-Real through Scene Reconstruction and Execution-Grounded Robot PoliciesLLM agents and tool use → Multimodal and embodied agents
  48. Before They Can Solve: Predicting Post-Training Coding-Agent Performance from Base ModelsLLM agents and tool use → Agent skill learning and evolution
  49. A Society of Researchers: Designing Institutions for Populations of Autonomous Research AgentsLLM agents and tool use → General LLM agent frameworks and benchmarks
  50. MORCA: Offline-to-Online Reinforcement Learning for Adaptive Cache Reuse in Video Diffusion AccelerationDiffusion models → Video generation
  51. PHRBench: A Behavioral Evaluation of Post-Hallucination Reasoning in LLMsLLM reasoning → Hallucination detection and mitigation
  52. RFPO: Rectified Flow Policy Optimization for Embodied ControlDiffusion models → Diffusion and flow policies for robotics
  53. Detecting Adversarial Images through Response Profiles of Vision-Language ModelsMultimodal LLMs / vision-language models → MLLM jailbreaks and adversarial robustness
  54. Conditional Flow Matching for Generation of 3D Multi-variable Instantaneous Urban Microclimate FieldsDiffusion models → Flow matching, time series and science
  55. Training Parallel Speculative Draft Models by Directly Minimizing Expected Decoding RoundsEfficient LLM inference and serving → Speculative decoding
  56. SOTA: Stock Options Trading Agents Guided by Option-Implied Return DistributionsLLM agents and tool use → Financial trading and forecasting agents
  57. Reasoning-Token Spikes Under Prompted Untruthful Responding in Large Language ModelsLLM reasoning → Prompting and chain-of-thought reasoning
  58. Self-correction Optimization for Interleaved Multimodal GenerationMultimodal LLMs / vision-language models → Image editing and quality assessment
  59. Explicit Geometric Chain-of-Thought for Vision-Language-Action in Autonomous DrivingLLM reasoning → Chain-of-thought reasoning for driving +2
  60. RoboQuest: Generalist Physical Agents that Search, Inspect and TestLLM agents and tool use → Multimodal and embodied agents
  61. ResidualQuant: KV Cache Quantization for Looped Transformers with 2-Bit ResidualsEfficient LLM inference and serving → KV cache quantization
  62. TaoD2C-Bench: Benchmarking MLLMs for Industrial UI Code Generation Beyond Visual FidelityMultimodal LLMs / vision-language models → Chart understanding and anomaly detection
  63. Koopman Observers for Diffusion Acceleration: Correcting Feature Forecasts with Shallow MeasurementsDiffusion models → Diffusion foundations and sampling
  64. Open-MMUnlearning: Unifying Methods and Evaluation for MLLM UnlearningMultimodal LLMs / vision-language models → MLLM jailbreaks and adversarial robustness +1
  65. SLDR: Defending Against Malicious Fine-tuning via Selective Layers Recovery and Dynamic RoutingLLM alignment / RLHF / preference optimisation → Safety alignment and jailbreak attacks
  66. Real-Time Joint Audio-Video Generation by Parallel Adapter CompositionDiffusion models → Audio, speech and music
  67. HAN-Mamba: Hierarchical Selective State Space Networks for Multi-Scale Financial Volatility ForecastingState-space sequence models (S4, Mamba) → Time series forecasting with SSMs
  68. SemanticFold: Latent Sequence Compression SeparatesLanguage Modeling, Decodability, and ReasoningLLM reasoning → Test-time compute and efficient thinking
  69. DeltaRepresentation: Geometry Supervised Representation Learning of Phenotypes via Counterfactual Reasoning for Medical VLMsLLM reasoning → Clinical and biomedical applications +1
  70. PatchBench: Measuring Collateral Damage in Activation PatchingLLM safety and security → Jailbreaking
  71. Sparse Planning in Visual World Models via Cost GradientsWorld models → Latent predictive world models
  72. Video Prediction Policy 2: Predict Better, Act BetterWorld models → Embodied and robot world models +1
  73. Geometry-Supervised Visual Representation Learning for Multi-Phenotype Lesion Interpretation in Medical VLMsMultimodal LLMs / vision-language models → Medical vision-language models
  74. Why Software Engineering Is Indispensable in the Age of Coding AgentsLLM agents and tool use → Software engineering coding agents
  75. How to train your model organismLLM safety and security → Safety alignment, refusal and guardrails
  76. Universal Local Error and Realized Amplification for the First-Order EDM PredictorDiffusion models → Diffusion foundations and sampling
  77. Kinetic Langevin Meets Split Gibbs: Accelerated Posterior Sampling for Imaging Inverse Problems with Diffusion PriorsDiffusion models → Inverse problems and image restoration
  78. Agentic AI-Assisted Modeling for Production Scheduling: Assessment in Constraint ProgrammingLLM agents and tool use → General LLM agent frameworks and benchmarks
  79. VideoEvolve: Co-Evolving Memory and Retrieval for Long Video UnderstandingLLM agents and tool use → Agent memory systems
  80. Beyond Outcome Rewards: Constructing and Assigning Retrieval Credit for Search AgentsLLM agents and tool use → Reinforcement learning for LLM agents
  81. Do Vision-Language-Action Models Understand Instructions? A Mechanistic Interpretability Study on Language GroundingVision-language-action models → VLA architectures and generalization
  82. Does Document Structure Help Dense Retrieval? A Placebo-Controlled Ablation of Four Mechanisms Across Two CorporaRetrieval-augmented generation → Generation and context integration
  83. UniSkill: Learning Actor-Aligned Skill Proposals for an Evolving PolicyLLM agents and tool use → Agent skill learning and evolution
  84. HeiCo-FOCUS: A Clinically Grounded Dataset for Long-Context Video UnderstandingMultimodal LLMs / vision-language models → Medical vision-language models
  85. HarnessIR: Harnessing Multimodal Foundation Models for Universal Real-World Image RestorationLLM agents and tool use → Multimodal and embodied agents
  86. When Algorithmic Exploration Becomes Cheap: A Case Study of Agentic Research in EDALLM agents and tool use → Educational agents and math problem solving
  87. Know the Shape, Find the Fault: Topology-Conditioned Diagnosis of Multi-Agent LLM FailuresLLM agents and tool use → Software engineering coding agents
  88. RealtimeWAM: How Fast Can I Run My World Action Model?World models → Embodied and robot world models
  89. HGP:An on-device personalized agent memory via hybrid graph storageLLM agents and tool use → Agent memory systems
  90. From Pixel to Coding: Evaluating the Figure Reproduction Capabilities of MLLMsMultimodal LLMs / vision-language models → MLLM training and visual instruction tuning
  91. Loud Failures, Quiet Failures: Fault Detection and Recovery in Tool-Using Language Model AgentsLLM agents and tool use → Tool learning and function calling
  92. The Long Road to the Same Answer: Cognitive Bias Under Escalating Reasoning Budgets in Large Language ModelsLLM reasoning → Reasoning benchmarks and evaluation
  93. Efficient Patch-Based Anomaly Detection Fused with Diffusion Driven Generative Modeling for Semiconductor Wafer Bin Map Open Set Anomaly DetectionDiffusion models → Flow matching, time series and science
  94. From Expected Harmfulness to Likelihood: A Probabilistic Reformulation of Jailbreaking LLM AgentsLLM agents and tool use → Agent threats, guardrails and governance +1
  95. Successive Training Stages and Large Language Model Persuasion: Effects of Misalignment, Supervised Fine-Tuning, and Preference OptimizationLLM alignment / RLHF / preference optimisation → Human values and AI feedback alignment
  96. Diagnosing Diversity Collapse and Validating Mask-Conditioned Diffusion for Labeled Microtubule MicroscopyDiffusion models → Medical imaging and segmentation
  97. AgentTime: Can Agents Estimate and Control Their Own Runtime?LLM agents and tool use → General LLM agent frameworks and benchmarks
  98. Many Ways to Succeed: Diversity-Driven RL Fine-Tuning for VLA GeneralizationVision-language-action models → Reinforcement learning and post-training
  99. Purifying Backdoored Large Vision-Language Models by Removing Hijacked DirectionsMultimodal LLMs / vision-language models → MLLM jailbreaks and adversarial robustness +1
  100. Juno: Taming Predictive Latents for Vision-Language-Action ModelsVision-language-action models → World model-based VLA
All 669
  1. RollVerify: Bridging Efficiency and Accuracy in Long-Tail Rollout Reinforcement LearningLLM reasoning → Reinforcement learning for reasoning
  2. FedSSMCoOp: SSM Encoders for light-weight Federated Prompt Learning for Few-shot ClassificationMultimodal LLMs / vision-language models → Prompt learning and adaptation of CLIP +1
  3. Constrained-Action AI Remediation for SIEM/XDR via a NeMo-Guardrails ProxyLLM safety and security → Prompt injection and agent security
  4. Defensive Sufficiency in a Stackelberg Model of AI SecurityLLM safety and security → Prompt injection and agent security
  5. A Scoping Review and Experimental Study on Reinforcement Learning from Human Feedback for Human-Robot CollaborationLLM alignment / RLHF / preference optimisation → Reward modeling and RLHF
  6. Think Before You Paint: Recursive Latent Reasoning for Diffusion ModelsLLM reasoning → Diffusion reasoning and test-time scaling +1
  7. LiveMACE: Process-Aware Evaluation of LLM Agent Capabilities in Evolving MarketsLLM agents and tool use → Financial trading and forecasting agents
  8. Training Advisors for LLM Agents from Task OutcomesLLM agents and tool use → Reinforcement learning for LLM agents
  9. Self-Evolve With a Reference:Anchored Training of Tool-Integrated AgentsLLM agents and tool use → Agent skill learning and evolution
  10. ORCA: Hunting Compositional Failures in Text-to-Image DiffusionDiffusion models → Image generation, text-to-image and editing
  11. SkillForge: Co-Evolving Skills and Agents via Dynamic Skill LifecyclesLLM agents and tool use → Agent skill learning and evolution
  12. Dual-QK: Sharp Queries and Flat Keys for Prunable 2-bit KV CachesEfficient LLM inference and serving → KV cache quantization
  13. Backdooring Acoustic Foundation Models for Physically Realizable TriggersLLM safety and security → Audio jailbreaks and speech backdoors
  14. BoT-GRPO: Efficient Process-Reward RL for Reasoning via Bag-of-Token AggregationLLM reasoning → Reinforcement learning for reasoning +1
  15. Judging in Latent Space: Efficient Generative Reward Modeling via Semantics-Preserving CompressionLLM alignment / RLHF / preference optimisation → Process reward models for reasoning
  16. UltraWorld: Learning Interactive Ultrasound World Models from Untracked Clinical Videos with Acoustic Sampling MapWorld models → Medical and surgical world models
  17. Relational Abstractions for Spatial Reasoning with Diffusion ModelsLLM reasoning → Spatial and embodied reasoning +1
  18. Decoupling Logic from Persona: Structural Immunity of Edge LLM Agents to Context PollutionLLM agents and tool use → Social simulation and multi-agent debate
  19. Beyond Policy Support: Interaction Constrained Offline Reinforcement Learning for Autonomous DrivingEnd-to-end autonomous driving → Imitation-learning end-to-end planners +1
  20. Unrolled Flow Models for ReasoningLLM reasoning → Test-time compute and efficient thinking +1
  21. EntroPrefill: Renyi-Guided Context Pruning with Conditional Stability Guarantees for Retrieval-Augmented GenerationRetrieval-augmented generation → Generation and context integration
  22. Diffusion-Generated Image Watermarking: A Two-Axis Taxonomy and Three Protocol-Bounded Case StudiesLLM safety and security → Watermarking generated content and models
  23. Flow-of-Thought: A Framework for Visual ReasoningLLM reasoning → Spatial and embodied reasoning
  24. DeltaWAM: Distilling Action Tangent Fields into World Action ModelsWorld models → Embodied and robot world models
  25. Bridge Routing Heads: Where Multilingual Multi-hop Reasoning Lives in LLMsLLM reasoning → Reasoning benchmarks and evaluation
  26. MeshCarve: Artisan Mesh Generation with Flow Matching in Compact Latent SpacesDiffusion models → 3D, point cloud and scene generation
  27. YUBI-STAG: Contact and Semantic-Rich Alignment for VLAs via Automated Video-Language GroundingVision-language-action models → Human-video and latent-action pretraining
  28. EC-EarthFlow: Probabilistic emulation of daily transient global climate model simulations with flow matchingDiffusion models → Flow matching, time series and science
  29. Understanding and Mitigating Token-Pruning-Induced Vulnerabilities in VLMsMultimodal LLMs / vision-language models → MLLM jailbreaks and adversarial robustness
  30. Latent Watermarks under Generative Editing: A Benchmark and Analysis of Detection SurvivalLLM safety and security → Watermarking generated content and models
  31. What Makes Synthetic Hard Negatives Work in Vision-Language Pretraining?Multimodal LLMs / vision-language models → Prompt learning and adaptation of CLIP
  32. Do Better Visual Representations Always Lead to Better End-to-End Autonomous Driving?End-to-end autonomous driving → End-to-end system architecture
  33. From Pareto to Preference: Personalized Test-Time Scaling via Amortized Agentic Policy DiscoveryLLM reasoning → Reasoning for agents, planning and code +1
  34. System Switch: When Should a Fast Decision Model Stop and Think?LLM reasoning → Reasoning for agents, planning and code
  35. Cost-Efficient Theorem Proving via Agent Orchestration in Program VerificationLLM agents and tool use → Software engineering coding agents
  36. InsClaimBench: Benchmarking Insurance Claim Adjudication Across the Decision ChainLLM reasoning → Reasoning for agents, planning and code
  37. Shared and structured inputs undermine collective random choice by reasoning AI agentsLLM agents and tool use → Social simulation and multi-agent debate
  38. Closed-Form Noise Calibration Against Membership Inference for Random-Allocation DP-SGDLLM safety and security → Membership inference attacks
  39. CircuitATLAS: Agentic reasoning over a systems neuroscience knowledge graph for target discovery in circuitopathiesLLM reasoning → Clinical and biomedical applications +1
  40. Coding-Agent Benchmarks Should Match Their Users' Task FlowsLLM agents and tool use → Software engineering coding agents
  41. How Do Agentic LLMs Decide to Call Tools? A Tool-Call Vector Shaped by SuppressionLLM agents and tool use → Tool learning and function calling
  42. Which Language Should a Skeleton Speak? Language Choices in Multilingual ReasoningLLM reasoning → Mathematical reasoning
  43. SafeEvo: Deciphering the Safety Alignment Mechanism and Evolution in Language ModelsLLM alignment / RLHF / preference optimisation → Safety alignment and jailbreak attacks +1
  44. Learning Situation-Conditioned Thinking Policies for Long-Term LLM AgentsLLM reasoning → Test-time compute and efficient thinking +1
  45. Collaborative Reasoning Distillation via Cross-Feedback and Coherent CurationLLM reasoning → Prompting and chain-of-thought reasoning
  46. Targeted Modality Dropout for Real-Robot Manipulation Robust to Intermittent Vision LossOffline RL and imitation learning from data → Robot manipulation imitation learning
  47. Visual Evidence Under Cross-Examination: Evaluating and Controlling Decision-Level Evidence Use in Vision-Language ModelsMultimodal LLMs / vision-language models → VLM benchmarks and visual reasoning
  48. Contact-Aware Imitation Learning Through Contact FactorizationOffline RL and imitation learning from data → Robot manipulation imitation learning
  49. Not All Uncertainty Matters: Simulation-in-the-Loop Fast-Slow Reasoning for Decision-Critical Autonomous Driving SystemLLM reasoning → Chain-of-thought reasoning for driving +2
  50. It Is Not Seeing the Hazard: A Frozen Vision-Language Safety Score Measures Its Caption BankMultimodal LLMs / vision-language models → MLLM jailbreaks and adversarial robustness
  51. STRIKE: Learning Visual State Transitions for Physical World ModelingWorld models → Video-generation world models and simulators
  52. Sparse Feature Policy Unlearning Mitigates State Hallucination in Vision-Language-Action ModelsMultimodal LLMs / vision-language models → Hallucination mitigation in LVLMs +2
  53. The Attribution Blind Spot: Layerwise Trajectory Diagnostics for Source Reliance in Retrieval-Augmented Language ModelsRetrieval-augmented generation → Agentic RAG, agents and memory
  54. Goldsmith: Gold-Loss-Guided Definition Optimization with an Agentic Annotation HarnessLLM agents and tool use → Agent skill learning and evolution
  55. MIMESIS: Learning User Simulators as Training Environments for Interactive AgentsLLM agents and tool use → Reinforcement learning for LLM agents
  56. Secure-CUA: Controlling Untrusted Influence in Computer-Use AgentsLLM agents and tool use → Agent threats, guardrails and governance
  57. LLM-Enabled UAV Dispatch: A System-Level Survey and TaxonomyLLM reasoning → Reasoning for agents, planning and code
  58. TMT: Runtime Backdoor Detection for Vision-Language-Action Policies on Unseen TasksVision-language-action models → Safety and adversarial attacks
  59. GeoPrior-Mamba: Structured Process Priors with Mamba for Fine-Resolution XCO2 ReconstructionState-space sequence models (S4, Mamba) → Time series forecasting with SSMs
  60. TempoBridge: Language-Guided Tempo Control for Vision-Language-Action PoliciesVision-language-action models → Reasoning and chain-of-thought in VLA
  61. Iris-3B: Going Beyond the Latent with Pixel-Space Diffusion Training, Conversion and Fine-TuningDiffusion models → Inverse problems and image restoration
  62. Beyond Token Revision: Investigating Mask-and-Replace Diffusion for Zero-Shot Text-to-SpeechDiffusion models → Discrete and text diffusion
  63. Mixture of Layers: Dynamic Layer Routing for Visual ReasoningLLM reasoning → Multimodal reasoning
  64. Controllable Crowd Generation through World-Model PlanningWorld models → Embodied and robot world models
  65. Event-Aligned Visual Action Reasoning for World Action ModelsWorld models → Embodied and robot world models +1
  66. Democratizing MoE inference on commodity GPUs with CoMoEMixture-of-experts → Efficient MoE training and inference systems
  67. Efficient Reasoning with Flow Language ModelsLLM reasoning → Test-time compute and efficient thinking +1
  68. Finding the Right Balance: Relevance and Diversity in LLM RetrievalRetrieval-augmented generation → RAG robustness and adaptive retrieval
  69. Noise, Denoise, Correct: MCMC Posterior Sampling with Diffusion Priors in Three StepsDiffusion models → Inverse problems and image restoration
  70. One Frame, Full Heartbeat: ECG-Free Cardiac Cine MRI Synthesis via Phase-Conditioned Flow MatchingDiffusion models → Medical imaging and segmentation
  71. From Plausible Hierarchies to Useful Taxonomies: Evaluating Agentic Harnesses on Customer FeedbackLLM agents and tool use → Web, search, and research agents
  72. Unified Multi-plane Autoregressive Diffusion for 3D Multi-contrast MRI SynthesisDiffusion models → Medical imaging and segmentation
  73. From Retrieval to Customer Context: Evaluating Frontier-Model Systems for Voice-of-Customer AnalysisLLM agents and tool use → Social simulation and multi-agent debate
  74. DUDA-Bench: Benchmarking LLM Agents on Multimodal Data-Driven Urban DiagnosisLLM agents and tool use → Multimodal and embodied agents
  75. Expert Coupling in MoE Pretraining: Reducing All-to-All Overhead with Correlated Placement and Token ShufflingMixture-of-experts → Efficient MoE training and inference systems
  76. Immiscible Diffusion Policy: Preserving Multimodal Robot Actions through Label-Free Noise AssignmentDiffusion models → Diffusion and flow policies for robotics
  77. From Chunks to Functional Evidence: Function-Aware Retrieval for EDA Documentation QARetrieval-augmented generation → RAG pipeline design and evaluation
  78. Multimodal LLMs Can Learn to Read Brain Signals: A Vision--Language Model for Unified Multi-Task EEG DecodingMultimodal LLMs / vision-language models → MLLM training and visual instruction tuning
  79. Learning Unknown Constraints without Unsafe Data via Optimality and Counterfactual RegularizationOffline RL and imitation learning from data → Inverse reinforcement learning
  80. Relevance Is Not Sufficiency: What Actually Closes the Evidence Gap in Long-Term Memory QALLM agents and tool use → Agent memory systems
  81. Shared Low-rank Basis Factorization for Data-free Mixture-of-Experts CompressionMixture-of-experts → MoE large language models
  82. SearchWorld: Spatial Value-Grounded Imagination for UAV Object Search via World ModelsWorld models → World models for navigation
  83. Constrained Diffusion for Data-Scarce Orbital Monte Carlo in Constellation TaskingDiffusion models → Diffusion and flow policies for robotics
  84. Why VLMs Miss Small Objects, and When Zooming In Is SafeMultimodal LLMs / vision-language models → Visual token pruning and efficiency
  85. Denoising Blocks, Not Tokens: Efficient Compressed Continuous Diffusion with Branching Token RealizationDiffusion models → Discrete and text diffusion
  86. Predicted Futures Are Not Enough: Learning Executable Goals for Robot ManipulationWorld models → Embodied and robot world models
  87. Kuration SDK: Addressing the Virtual2Real Gap via Data CurationWorld models → Video-generation world models and simulators
  88. RT-Safe: Benchmarking Agent Safety in Real-Time Embodied EnvironmentLLM agents and tool use → Agent threats, guardrails and governance
  89. LeCuration: A Tiny World Model as a Data Curation Multi-ToolWorld models → Latent predictive world models
  90. Package Hallucination Attacks on Coding Agents through Prompt Injection in Rule FilesLLM agents and tool use → Agent threats, guardrails and governance +1
  91. Adaptive Visual Token Reduction for Accelerated Image UnderstandingMultimodal LLMs / vision-language models → Visual token pruning and efficiency +1
  92. Efficient Best-of-N policy evaluation for inference-time alignmentLLM alignment / RLHF / preference optimisation → LLM preference optimization methods
  93. We Query, Therefore We Compute: On Oracle Computation beyond the Machine, with an Application to AgentsLLM agents and tool use → Software engineering coding agents
  94. Adversarial Images Hijack Web Agents from Visual Grounding to Browser ExecutionLLM agents and tool use → Multimodal and embodied agents +1
  95. The Winner's Curse in LLM Self-Improvement Loops: Selection Noise, Lock-in, and Acceptance RulesSynthetic data and self-improvement for LLMs → Self-rewarding alignment and self-training
  96. Prompts versus Rules: Auditing and Controlling Speech Naturalness Behaviors in Voice User Simulators
  97. Distillation for Incrimination and Distillation for Capabilities
  98. PCAsplat: Gaussian Splatting with Local PCA Regularization
  99. Reconstruction of Multiscale Plasma Dynamics Across Operating Regimes
  100. Low-rank tensor structure of precipitation and its application to satellite-reference merging
  101. F3NO: Frequency-Decomposed Finite-Time Flow-map Neural Operators with Cross-Scale Conditioning
  102. Region-Aware CLS Token Augmentation for Fine-Grained Image Retrieval
  103. Omni-Diffusion-Distill: Few-Step Distillation of Unified Multimodal Diffusion Large Language Models
  104. Multi-Bandwidth Distribution Matching Distillation: On the Equivalence of Distribution Matching Distillation and Drifting Models
  105. Fluid-Gen-Zero: Grounding Pretrained Video Generators in Physics without Training
  106. Probabilistic Sensing, Deterministic Authority: Admitting Model-Produced Observations into Sufficiency-Checked Governance Contracts
  107. Optimizing Large Language Models with Chained LMOs
  108. Budgeted Multi-Source Counterfactual Annotation for Off-Policy Evaluation
  109. Transferability of Learned States in Neural PDE Solvers
  110. When Citations Mislead? A Claim-Level Benchmark for Legal Hallucination Detection
  111. ASPIRE: Saddle-Point Discovery through Set Prediction and Physical Refinement
  112. Spectrally Targeted Muon
  113. LVSPM: Long Sequence View Synthesis and Pose Estimation Model
  114. GPU-Accelerated Computation of Persistent Homology for Topological Analysis of Image Data
  115. TRACE: A Governance Framework for Measuring Explainability Debt in Production AI Systems
  116. Learning How to Search for Plans with Exponentially Less Space
  117. SPD-MetaFormer is what you need for small-data brain decoding
  118. AI4Fire: Evaluating Large Language Models on Wildfire Tasks
  119. GHARP: Real-time Gaussian Head Animation from Large-scale Reconstruction Prior
  120. Higher-Order Morphology Priors for Quadruped Reinforcement Learning Under Actuator Degradation
  121. Rethinking the Tradeoff Between Temporal Encoding and Nonlinear Computation in Spiking Language Models
  122. Coefficient Calibration as Selection Pressure in Symbolic Regression
  123. Adaptive Multi-Discriminator WGAN Framework for Resource-Constrained Internet of Vehicles Using Reinforcement Learning and Game Theory
  124. The Missing Fourth Term for the Emulation Tensor Memory Equilibrium (TME) Model: The Residue Deconstruction Cost
  125. Language Models for Page-Level Layout Decisions in E-commerce Search
  126. Implementation Guidelines for Data Quality Metrics
  127. Large Language Models for Machine Translation Quality Annotation: Humans and Models Are Both Challenged
  128. When Listening Becomes Easier: Scrubbing Visual Cues for Shortcut-Free VLAs
  129. Reading the Room: Foundations, Design, and Challenges of Normative Competence in LLMs
  130. How Hackable Is Your Speech Quality Metric? A Corrected Protocol, a Benchmark, and What Patching Buys
  131. Gen-PINNs: Generative Adversarial Physics Informed Neural Networks for solving partial differential equations
  132. Barron Optimal Transport I: Generative Modeling
  133. Sparse Attention Is Matrix Approximation, Not Choosing from a Bag of Values
  134. Transformed Samplers with Variance Reduction
  135. Conversational Voice Aesthetic Model with Reinforcement Learning from Human Listeners
  136. Shape irregularity of Life-Like Network Automaton rules as an indicator of classification performance
  137. Disentangling Linguistic and Paralinguistic Information with Routed Sparse Autoencoders
  138. A mesh-based neural energy method for the simulation of heterogeneous composites
  139. Self-Supervised Keyframe Discovery for Horizon-Invariant Behavior Cloning
  140. KDFP: A first-principles approach to knowledge distillation in large language models
  141. Similar Predictive Fit but Different Latent Dynamics: Characterizing Learned Dynamical Structure in Personalized Models of Brain Disorders
  142. Amortized Off-Policy Evaluation for LLMs
  143. Real Long-Term Memory for AI: A 50-Million-Token Window That Is Faster and Cheaper Than Recompute
  144. On the Clock: Towards Punctual and Productive Time-Budgeted AI Agents
  145. MotherTree: Meta-learning on synthetic data improves decision tree training
  146. Conformal Prediction under Partial Verification
  147. Grammar Concept Annotation at Scale: Deployed Fine-Tuned Small Language Models Outperform Prompted Frontier Models
  148. Diagnosing and Recovering from Observation-Space Shift at Long-Horizon Skill Seams
  149. Evaluating Rubric Generation with Interventional Transfer
  150. Controlled Acquisition and Abstention in Three-Channel Score Conflicts
  151. Symbolic Density Estimators for Unnormalized Distributions
  152. Whose Ground Truth? Embracing Ambiguity in Human-Centered AI
  153. How Many Repeated Pairwise Comparisons Are Needed for Ranking under Heterogeneity?
  154. Calibrating Ambiguity Set via Diagnostic Transport for Distributionally Robust Optimization
  155. NEMORA: Neural Equivariant Multipole Operators for Long-Range Atomistic Learning
  156. AI-Mediated Self: How HCI Defines and Relates to the Self
  157. Strategic Investment Decision Making for Value Creation in Energy Transition: A Reinforcement Learning Approach
  158. CPU-Auth: Device Fingerprinting for Authentication via DVFS Side-Channel
  159. What can linear attention learn from nonlinear teachers in-context?
  160. DOGS: Design-Space Sampling for Prompt-Driven Logo Generation
  161. MESSENGER: Memory-Enhanced Sequential Scene Flow Estimation via Autoregressive Next-Frame Forecasting
  162. Clarify, Then Focus: Statement Normalization for Conversation Analytics at Scale
  163. Conversational Task Disambiguation over Tabular Data: Leakage-Aware Formulation, Benchmark Suite, and Training
  164. Lossy Compressive Text Autoencoders
  165. Deep Learning vs. Statistical Models for Multi-Horizon Price Forecasting of Second-Hand Electronics: A Systematic Benchmark
  166. Cognitive Thermometers: Machine Learning and Logical Complexity
  167. LinSlot: Exploiting Linear Representation hypothesis for unsupervised attribute discovery from slot based object representation
  168. Seeing Through the Glare: A Multi-Source Benchmark and Ocular-Adaptive Pixel MeanFlow for Eyeglass Reflection Removal
  169. On linearity or non-linearity in machine learning for quantum chaotic dynamics
  170. Learning infinite context windows in recurrent architectures via spatial neural computing
  171. Explaining the Saliency Map Sparsity of Adversarially-Trained Neural Networks
  172. BEANS-Next and ROOTS: Broadening Audio-Language Capabilities for Bioacoustics
  173. Teaching PPG How not Who: Fixed-Effects Distillation from ECG
  174. Beyond Owls: Subliminal Learning Can Transfer Learned Capabilities and Backdoors
  175. PXtal: Learning to Align Powder X-Ray Diffraction and Crystal Structures under Information Asymmetry across Modalities
  176. Beyond the Ergodic Wall: A Discrete Geometric Physics Sandbox for Analysing AI Scaling Limits and Complexity Collapse
  177. Large Language Model-Assisted Preparation of Transportation Management Plans: A Case Study with WisDOT WisTMP System
  178. Leakage-Controlled Multimodal Learning for Diagnosis and Progression Prediction in Alzheimer's Disease Research
  179. Masked Generative Motion Planning with Geometry-Guided Token Search
  180. From Log-Odds to Shapley Values: An Explanatory Geometry for the Weighted Naive Bayes Classifier
  181. D-SLR: The Disjoint Row-Sparse plus Low-Rank Decomposition
  182. Has LLM Screening Performance Stalled in Software Engineering Systematic Reviews?
  183. Phase-HDC: Replacing Optimizer History with Gradient Thresholds in Discrete Phase Learning
  184. Sample-Efficiency of Kolmogorov-Arnold Networks
  185. Exact SO(3)-Equivariant Isotropic Kernels for Rotation-Robust Neural Dynamics
  186. SAVU-BENCH: A Real-World Benchmark for Spatial Audio-Visual Understanding
  187. WorldBench: Evaluating LLMs on Three.js Voxel World Generation
  188. Self-Organization from Constrained Geometric Radiation
  189. When AI Finds Hidden Messages, Does It Report?
  190. MRCert: Towards Post-deployment Patch Robustness Certification for Adversarially Patched Samples via Type-specific Masking
  191. JevForest: Path Voting for Budgeted Feature Acquisition
  192. Temporal transformer CAN encoder with federated lightweight heads for anomaly detection
  193. A Camera-Native Stereo VR180 Dataset
  194. Tetris3D: 3D Scene Generation With Objects That Fit Together
  195. Never Look Back: Understanding Persistence in 3D Object Memory from Egocentric Videos
  196. Decoupling Exploration from Optimization in RLVR
  197. EngramEdit: Decoupled Knowledge Updates in LLMs through Conditional Memory
  198. Decentralized SGD under Heavy-Tailed Noise: Optimal Convergence Rates and the Role of Gradient Clipping
  199. Distilling Graph Geometry: Knowledge Gap from GNNs to MLPs
  200. SciExam for ENSO: Can AI Agents Build Climate Models?
  201. Video-Conditioned Generative Joint 2D-3D Hand Motion Recovery
  202. Your Prompt Should Do More: Effects of Retrieval Instructions in Embedding Models
  203. RECAST: Learning to Compute the Right Context through Adaptive Evidence Routing
  204. Validity Without Ground Truth: What Stated-Preference Economics Offers the Evaluation of Language Models
  205. Oracle-Efficient and Parameter-Free Agnostic Smoothed Online Learning
  206. EmbodiedRSI: Active Continual Robot Learning Through Hypothesis-Guided Co-Evolution
  207. QuadTok: Quadtree Visual Tokenizer for Autoregressive Image Generation
  208. Evolutionary Architecture Search for Chlorophyll-a Prediction in Lakes using Sentinel-2
  209. Insights from Autoresearch for Solar Panel Segmentation
  210. Best Arm Identification for Bandits with Shifting Means
  211. Two-Level Softmax Sampling Done Right: Correcting Bias from Size Imbalance and Dispersion
  212. Label-free cell counting and viability prediction with brightfield imaging and deep learning
  213. How assigned AI use before class shapes active student engagement in class
  214. FoldBack: Self-Correcting Masked Generative Policy for Long-Horizon Garment Folding
  215. Composing What Each Teacher Learned: Multi-Teacher On-Policy Distillation through Teacher-Relative Shifts
  216. NeuralBES: A Differentiable, Control-Aware Emulator for Scalable Building Energy Modeling
  217. MemoCare: An Interactive Multimodal Mobile System for Automated Cognitive Screening
  218. A Good Self-Teacher Meets the Student Where They Are: Joint On-Policy Learning and Teaching
  219. RunningTab: Direct Workspace Interaction with Environment-Side Tabs
  220. Seq-Flow: Efficient Probabilistic Forecasting with Self-Rollout Error Control
  221. ECHO: Embodied Camera Observations of Human Object Carrying
  222. Q-Learning with Scalar Adjoint Matching
  223. SGF+: Decoupling Gradient Flows for Autoregressive Video Generation
  224. Derivative Gaussian Processes on a Two-Direction Budget
  225. CoTrace: Data Recipes for Training Terminal Agents with Harness-Model Co-Evolution
  226. GraphRectify: Graph-Based Transfer of Adversarial Example Detectors Across Neural Networks
  227. Which Rollout Taught It That? BehaviorTrace and the Limits of Training-Data Attribution in Online RL
  228. Steerspeech: Activation Steering For Emotion Control In Generated Speech
  229. RobotWorld: Benchmarking Multimodal Agents for Robot Use Across Diverse Tasks and Embodiments
  230. Rubix: Global Correspondence-Free Point Set Alignment through Assignment Geometry
  231. Cross-Domain Pretraining for Steady-State Neural CFD Surrogates
  232. Gaussian Density Splatting Network
  233. Executing Causal Structure Learning with Linear-Attention Transformers
  234. Kernel Autoresearch for Open-Ended Model Discovery
  235. Safe Meta-Policy Design with Risk Control
  236. MOTIP2: Spatial Priors for End-to-End Multi-Object Tracking
  237. OrBIT: Structure-Guided Embedding Compression
  238. Boosting and the Expressive Power of Simple Weak Learners via the γ-VC Dimension
  239. Continual Learning without Continual Training
  240. Document-Level Text Simplification in Estonian Using Large Language Models
  241. PalmSpace: Towards a Versatile On-Palm Interaction Space through Unified Touch Modeling
  242. Input-Blind Controls Produce Substantial Oracle Headroom for Layer Programs in Multiple-Choice Evaluation
  243. Temporally Interpretable Differentiable Decision Trees
  244. Pathwise Information Certificates for Decentralized Adaptive Sensing
  245. ORDERS: An Empirical Study of Norm-Rank Aggregation for Personalized Federated Learning
  246. MultiFly: A Real-World Multimodal Aerial Dataset with Annotation-Efficient Label Transfer and Cross-Modal Semantic Consistency
  247. MIRA: A Musical Intent Refinement Agent for Aligning Text-to-Music Generation with User Intent
  248. The Handover Problem: Governing Autonomy Transitions in Human-AI Collaboration
  249. Measurement-Efficient Differentiable Quantum Architecture Search for Combinatorial Optimization
  250. AutoAdapt: Automatic Domain Discovery Enables Low-Cost Extensibility
  251. Dataset Pruning from First Principles: A Label-Free Linear Programming Approach
  252. Position Forcing: Self-Conditioning 3D Generation
  253. Data Reuse in Non-Stationary Learning
  254. When to Unpair: Regulating Pairing Dependence in Medical Visual In-Context Learning
  255. How Private is Private? A Comparative Study for Face De-Identification
  256. Learning to Act with Task Progress: Distilling Small Agents from Compact Teacher Supervision
  257. Average-Reward Reinforcement Learning for Multichain MDPs: A Hierarchical Decomposition Approach
  258. Performance at What Cost? A Sustainability-Aware Performance Index for Cell and Nucleus Instance Segmentation
  259. From Digital Human Interactions to Physics-Based Humanoid Skills: Physics-Grounded Post-Training of Interaction Generators
  260. Estimating Uncoded Crash Factors with Tabular Foundation and System One Models: Kumo Tabular and Jev
  261. LLM-Assisted Generation of Transparent, Open-Source Multiphysics Models of Electrochemical Devices
  262. Nobody Truly Agrees on Sentiment: Humans, Bespoke Tools, and LLMs Struggle with Social Media Texts
  263. Thinking in Depth: Retrospective Inference for Tabular Foundation Models
  264. PoreML: A Data-Driven Framework for Learning Multiphase Flow in Porous Media
  265. Fault-tolerant foundation models
  266. RSIGym: A Flexible Environment for Recursive Self-Improvement
  267. One-Shot Adaptive Segmentation For Scientific Images
  268. How Do Transformers Learn to Represent Symmetries?
  269. On the Necessity of Attention-FFN Split in Vision Transformers
  270. Continual Graph Multi-Agent Reinforcement Learning
  271. Revisiting Explainable AI through Model-Independent Concept Dictionaries
  272. Shared Gaussianization: What Gaussian Regularizers Certify About Contrastive Learning, and What They Miss
  273. Physics-Aligned Electronic Ground-State Learning Improves Generalization
  274. Energy-Efficient Gait Adaptation via Hierarchical Reinforcement Learning for Quadrupedal Locomotion Across Diverse Terrains
  275. TouchScale: 500 Hours of Human Vision and Touch for Visual-Tactile Learning
  276. AI Safety Considerations for Agents With Limited Time to Act
  277. Temporal Visuo-Tactile Learning for Dexterous Grasp Stability
  278. Neural Sampling with Reweighted Normalizing Flows via the Wasserstein--Fisher--Rao JKO Scheme
  279. A Closed-Loop Non-Asymptotic Convergence Analysis of PPO with Learned Critics and Clipping
  280. LoomSC: Scalable Deep Subspace Clustering with Projector Factorization and Exact Spectral Reduction
  281. Stale, Misattributed, or Late: Where Personal Memory Fails Before Generation
  282. Using Small Language Models to Reverse-Engineer Machine Learning Pipelines Structures
  283. PairAudit: Guiding Human Review with Graph Tokens under Distribution Shift
  284. QuSema: Detecting Silent Bugs in Quantum Libraries via Quantum-knowledge-enhanced Agents
  285. OOM-RL II: Reality Is an Oracle, Not a Debugger Provenance-Constrained Diagnosis in Continually Evolving Agent-Engineered Systems
  286. On the Cyclic Assumption of the Cow-Path Search Algorithm
  287. Logarithmic Regret via Passive Change Detection in Piecewise-Stationary Self-Tuning Regulation
  288. Stationary Bias and Extrapolation in Nonlinear Two-Timescale Stochastic Approximation
  289. MorphCL: Morphological Contrastive Learning for Inertial-based Human Activity Recognition
  290. TestGRAD: Evolving Test Suites via Failure Pattern Momentum for SWE-Agent Ensemble
  291. When Scientific Cognition Is No Longer Scarce
  292. ProtocolMatch: Protocol-Dependent Model Selection for Scientific Dynamics Forecasting
  293. Beyond LLM-GA: Secure Fluid Antenna Systems with ReEvo-Designed Memetic Algorithm
  294. LLM Persuasion Is in the Eye of the Evaluation
  295. From Prompts to Trees: Effective LLM-Guided Tree Generation for Few-Shot Tabular Classification
  296. Masked Feature Encoding for Large-Scale Whole Slide Image Representation
  297. Evaluating Sequence Assembly Strategies for Differentially Private Synthetic Time-Series Forecasting
  298. Finite-Sample Approximation of Hessian-Guided Perturbed Wasserstein Gradient Flows
  299. RoBART: Bayesian Additive Regression Trees with Tree-Specific Rotations
  300. Edge Accuracy Is Not Enough: Why Dynamics-Learned Structure Fails to Transfer to Inverse Problems
  301. OrthoGen: A Generative Orthogonal Learner for Time-Varying Treatments
  302. CARES: A Controlled Synthetic Benchmark of Speaker Reactions to Sound
  303. Computations of the slice genus and the unknotting number of links via machine learning
  304. GAGR-Lab: Evaluating Joint Spatial-Geometric and Analytic Function Reasoning
  305. Robust Decentralized Fairness Auditing
  306. VolCo: Volumetric Contact for High-Fidelity Human Grasp Generation
  307. HuLiGen: Human LiDAR Generation from Parametric Body Models
  308. Broadly Applicable Approximate MCMC for Switching Stochastic Differential Equations Using Uniformization and Time-Conditioned Factorized Neural Likelihood Estimation
  309. Pre-training of Bayesian Optimization Algorithm through Bayesian Optimization
  310. EEG and Eye-Tracking Evidence That AI Disclosure Shapes Face Evaluation
  311. Argos: Adapt Rich Geometric Priors for Generalizable Online Scene-Change-Detection
  312. A Unified Information-Theoretic Approach to Constrained Multi-Fidelity Multi-Objective Bayesian Optimization
  313. Conformal Prediction for Spatially Dependent Data via Sequential Whitening
  314. Policy Learning with Weak Signals
  315. Beyond Anonymous Captions: Grounding Character Identity in Video Captioning and Question Answering
  316. BagDINO: Multi-View Baggage Re-Identification with DINOv3
  317. HySPE: Positional Encoding via Symplectic Dual Shears
  318. InterView-C: A Synchronized Multimodal Corpus of VR Avatar-Mediated Survey Interviews
  319. Progress and Prospect of AI in ARPES Workflow
  320. LLM4Impact: Integrating Heterogeneous Information for Scientific Impact Prediction
  321. Attention via Black-Box Vector Search
  322. m-Set Adversarial Bandits with Winner Feedback
  323. Lifelong small-object navigation in changing object layouts: a benchmark and method
  324. Training with Missed Targets in Generative Recommendation: Separating Supervision from Probability Competition
  325. What Can a Gaussian Process Design Test
  326. RewardWeaver: Long-Horizon Interactive Learning for Language Agents via Self-Evolving Reward Adaptation
  327. YANchor-4B: Effective Long-Horizon Reasoning in O(N) Time with O(1) Memory
  328. A Probabilistic Perspective on Wasserstein-Based Evidential Uncertainty for Out-of-Distribution Segmentation
  329. Mechanics of Long-Context Hybrid Models Part 1.1: From Hybrid Attention to Hybrid Position
  330. CAFE+FNO: Fourier Kernel Generation via Multiplicative Feature Composition
  331. I would rather quit NLP than read another paper like this: The rise of antithesis in NLP papers
  332. ExperienceIndex: Artifact-Grounded Memory
  333. Temporal Residual Bottleneck for Robust Asynchronous Collaborative Perception
  334. SkillSandbox: Skill Verification via Dynamic Scenario Synthesis
  335. Multi-Agent Coordination via Support-Preserving Distillation
  336. Activation-Aware Weight Tensorization: A Calibration-Time Preconditioner for Tensor-Network LLM Compression
  337. Sharp Asymptotic Theory of Maximum Likelihood Estimation for Gaussian Processes with an RBF Kernel
  338. Towards Calibrated Probabilistic Forecasts for Events of Interest via Outcome-Conditional Recalibration
  339. Efficient Provably Private Classification with a Tabular Foundation Model
  340. Comprehension Audits to Mitigate Risks from Automated AI Research
  341. TRACK: Telemetry-Based Racing Analysis and Coaching Kit in Sim Racing Games
  342. Cache the Encoder Within:Compact, Reusable Memory across LLM Queries
  343. WxFM-XL: Adapting Univariate Foundation Models to Multi-Station Weather Forecasting
  344. Transition Path Sampling Using Koopman Operators and Exit-Time Optimal Control
  345. AdSpark: A Large-Scale Dataset and Benchmark for Product-Centric Advertisement Video Generation
  346. Learning to Accumulate Knowledge with Mutual Information
  347. Evolve on the Host, Predict on the Edge: Deploying Online Neuroevolutionary Architecture Search for Cross-sectional Stock Return Prediction
  348. Matching of signal, noise and hardware timescales for filtering and forecasting of correlated noise signals
  349. Gaussian Equivalence for Multi-Head Self-Attention
  350. Structure alone supports efficient visual computation in the Drosophila visual system
  351. Controlling Dependence in Implicit Generative Models via Spread Mutual Information
  352. Oscillatory Neural Dynamics over Sheaves
  353. What the Sleeve Feels: Explainable Machine Learning for Textile Pressure-Based Postural Screening
  354. A Drosophila Whole-Connectome Network Can Learn Human-Designed Cognitive Tasks
  355. Scalable Patch-Level Self-Supervised Learning
  356. Playing with Kruskal: algorithms for flat and hierarchical watershed cuts
  357. Perceptually Aligned Evaluation of Style Transfer
  358. Beyond Reward Suppression: Near-Optimal Offline Attacks on Warm-Start Bandits with Bounded Rewards
  359. Temporal Predictive Multiplicity: Equally Accurate Time Series Models Yield Different Forecast Trajectories
  360. Marrying Pricing and Advertising with LLMs
  361. Extreme Binary Classification: Extreme Value Theory for Extreme Constraint on False Negative
  362. Sensitive-Topic Leakage Through LLM Routing Metadata: Measurement and Mitigation
  363. EASE: Entropy-Adaptive Distribution Shaping for Evading AI-generated Text Detectors
  364. TR-PTQ: High-Accuracy Integer-Only Transformer Post Training Quantization via Taylor Region Reformulation
  365. Force without transmission: a depth-induced rank collapse that no loss on the representation reopens
  366. Possibilistic Radial Transport for Approximate IM Inference
  367. MSU Team at the Explainable Deepfake Detection Challenge 2026: Grounded Artifact Evidence for Deepfake Detection
  368. Learning Traffic Flow Dynamics with Stochastic Physics-Informed Neural Cellular Automata
  369. Fast holographic inversion of superconducting domes
  370. Where Can a Decision Model Diagnose HVAC Faults? Reasoning Demand, Physical Representation, and Robustness Under Shift
  371. Itgan at NADI 2026 shared task: Parameter-Efficient Whisper Adaptation for Robust, Mixed-Dialect and Code-Switched Arabic ASR
  372. Expected Sample Complexity in Multi-Armed Bandits
  373. Do Generative Priors Align with Human Naturalness Perception?
  374. KGATE : a Knowledge Graph Embedding Training Environment
  375. Inverting Multi-Vector Visual Document Indices
  376. Eigenvalues of the Hessian in Deep Learning: The Origin of Symmetry and Its Breaking
  377. NeuralZip: Reusable Setup for Fast Lossless Compression
  378. Learning joint probabilistic weather forecasts from station observations alone
  379. NL2Hull: A Natural Language-Driven Constrained Ship Design Decision Framework
  380. QCATS: Query Context-Aware Transformer Slicing for Efficient Predictive Query Processing
  381. Outperformance Inverse Optimization: Learning Objective Functions that Outperform Agent Decisions
  382. Identifiability of a dissipative knowledge-dynamics model: exact recovery under designed excitation, degeneration on observational data
  383. Sparsifying Stochasticity, Not Capacity: Partial Stochasticity via Deep Weight Factorization of Prior Scales
  384. Layerwise Error Attribution for Fast and Robust Mixed-Precision Post-Training Quantization
  385. SANet: Selective Attention Network for Infrared Small Target Detection
  386. Bringing BNNs to Fast Event Processing
  387. An AI-assisted conditioning and geological interpretation workflow for usage in implicit geological modeling
  388. Deadline-Aware Multi-Agent Reinforcement Learning for TSN-Based Vehicular Edge Networks
  389. MUNITE: Unified Multimodal Latent Inference for Any-to-Any Multimodal Generation
  390. Global Average Precision for Representation Learning
  391. DeepTopoClustering: Unsupervised Derivation of Surface Process Taxonomy from 4D Point Clouds for Topographic Monitoring
  392. DeltaSplat: Iterative Gaussian Refinement for Pose-Free Feed-Forward 3D Gaussian Splatting
  393. Hard, Yet Reducible: Controlled Forward Transfer for Synthetic Degradation Curation
  394. Stream-Based Active Learning with Cooperative Neural Networks for Data-Efficient Partial Inverse Design: An Automotive Glass Run Channel Case Study
  395. For Those Who Believe in Faithfulness: Optimizing the Area Under Insertion and Deletion Curves for Ranking Relative Feature Importance
  396. Fully Interpretable Minimal Transformers: From Geometry to Algorithm
  397. Origins of Universal Machine Learning Force-Field Errors in Multicomponent Materials
  398. A Deafening Silence: Catastrophic Forgetting Lives in the Output Embeddings of Tokens the Data Never Speaks
  399. MOTIF: Person-of-Interest Deepfake Detection Beyond 3DMM Coefficients
  400. Homogenization in Multi-Agent Systems
  401. UltraText Bench: A Comprehensive Bilingual Benchmark for Evaluating Visual Text Rendering in Image Generation
  402. Efficient 3D Gaussian Head Avatars for Edge Devices
  403. Concentration, Not Uncertainty: Why Targeted Synthetic Data Doesn't Help Camouflaged Object Detection
  404. DisParQ: Self-Supervised Part Concepts for Interpretable Vision Foundation Models
  405. Beyond Masks and Trajectories: Flow-Guided Latent Action Injection for Stable Surgical Video Generation
  406. MIRROR: From Imitation to Internalization in LLM Personalization
  407. A Proof-of-Concept Study of Weakly Supervised Labeling of Fine-Grained EEG Components for Artifact Attenuation
  408. Counterfactual Route Optimization for Gaussian Head Avatar Modeling
  409. CIRSeg: Coarse-to-Fine Intensity-Robust Liver Segmentation with Source-Free Continual Test-Time Adaptation
  410. AdaPS-LiNGAM: Adaptive Predecessor Selection for Linear Non-Gaussian Acyclic Models under Small-Sample Settings
  411. Reproducible LLM Inference Benchmarking: A Sequential Isolation Protocol for Regression Testing
  412. Artificial intelligence pathways from weather to climate
  413. From Expert-Guided Proof Search to Automated Open-Problem Solving
  414. Fluctuations of Nonlinear Observables in Mean Field Neural Network Training
  415. PARC-Loc: Text-to-Point-Cloud Localization with Partial Assignment and Relational Consistency
  416. Leaner Transformers Can Easily Learn to Cluster
  417. Shaer: Controlled Arabic Poetry Generation with Meter Subform and Semantic Conditioning
  418. SoftSEEPS improves ML-based precipitation forecasting
  419. SoccerNet-FoulRet: Retrieving Semantically Similar Soccer Foul Videos
  420. Healthy skepticism in AI: a data visualization research agenda
  421. A Strength-Monotonic Law for Domain Alignment in Frozen-Embedding Bioacoustic Classification
  422. Unbounded Characteristic and Universal Kernels
  423. Beyond Group Splits: Specimen-Level Cross-Validation and Visual Attribution for Remaining-Shelf-Life Regression in Climacteric Fruit
  424. Towards Explaining Query Expansion Performance in Information Retrieval
  425. DynStream: Online Streaming 4D Gaussian Reconstruction of Dynamic Worlds from Unposed Video
  426. Pareto-optimal quantum kernel selection for unsupervised anomaly detection on real malware beaconing data
  427. Boundary-aware Reinforcement Learning for Hypercube State Spaces via Deterministic Policy Gradient
  428. SpikingVLA: Asynchronous Spiking Vision-Language-Action Models
  429. Pretraining Shapes Spectral Structure: Architecture- and Strategy-Conditional Prediction of OOD Robustness in Foundation Models
  430. Enhancing Multi-Region Stylization with Interior-Guided Boundary Repair
  431. A Tale of Two Error Categories: Exploring Concealed Trade-Offs in the Errors of Automated Judges in Evaluation of Uncertainty Quantifiers
  432. Few-Shot Learning for Personalised Automated Pain Assessment
  433. Optimal Regret for Online Market Making with Limit Order Book
  434. NAViLoss: An Underwater Navigation-Aware Dual-Residual Objective for Physics-Consistent Learning
  435. The Silhouette Operator: Identifiability of Low-Rank Measures from One-Dimensional Projections
  436. CERO: Where and When to Allocate Rollouts for RL Post-Training
  437. A Multi-Source Ultrasound Benchmark Revealing the Limits of Contemporary Self-Supervised Anomaly Detection Methods
  438. Gauss-Newton Accuracy and Indefinite Hessians: Uniform Coexistence in Low-Cost Sets
  439. Quasi-Binarized Autoencoders: An Architecture-Independent Information Bottleneck for Medical Image Anomaly Detection
  440. SAPD: Step-Aligned Privileged Distillation
  441. Real-world application of deep learning in large-scale seismic interference attenuation: A case study in the Camie field of Angola
  442. Alice: A Large-Scale German Benchmark for Rubric-Based Multi-Dimensional Automatic Short Answer Scoring
  443. Rubric Spans are Label Representations: Joint LLM Encoding for Short Answer Scoring
  444. DSTNet: Dynamic Spectral Trajectory Network for Causal Multi-Horizon Financial Forecasting
  445. MeshSIPP: Efficient Lattice Planning in Dynamic Environment
  446. When Rank Rises as LLMs Degrade
  447. Automatically Building and Updating a Knowledge Graph of MLIP Models
  448. Q-PhotoMarket: A Design Space Exploration Framework for Photonic Hybrid Quantum Neural Networks in Financial Market Prediction
  449. On-Policy Distillation Teaches New Skills but Not New Knowledge
  450. Tracing Inputs, Verifying Outputs: Validating Attribution in Music Generation
  451. Quantum anomaly detection in real scarce data
  452. When does a network's training history predict its future learning better than its current state? Evidence from a response probe and a forecasting screen
  453. The Identifiability and Observability of Deep Normalized Attention
  454. Second-order optimization of variable projection SVM models and road abnormality detection
  455. Identity-Duplication Auditing in National-Scale Neuroimaging Repositories
  456. Residual Learning in Empirical Asset Pricing
  457. Sequential Pretraining Favors Large Models
  458. Decoupled Optimization for Teacher-Student Semi-Supervised Learning via a Pioneer Student
  459. Lightweight and Versatile Learned Optimization by Recombination of Gradient History
  460. Cluster-Robust Prediction-Powered Inference
  461. STORK: Spatio-Temporal Observation of uterine contRactions via neural networKs
  462. COPC: Coupled Off-Policy Correction for Asynchronous LLM Reinforcement Learning
  463. Structured pre-generation elicitation versus single-shot prompting in AI-assisted enterprise decision-making: a randomised online experiment
  464. Scalable Logistic Gaussian Process Density Regression with Kinetic Langevin Sampling
  465. Dual- versus Single-Suggestion AI Support for Radiographic Interpretation in Residents: Randomized Multireader Study
  466. Adaptive Code Generation for Controlling Robots
  467. Correct Answers, Unsupported Findings: Evidence Binding in Forensic Reconstruction of LLM Agent Logs
  468. Gradient-Based Trajectory Optimisation over Continuous Poses for Sparse-View Cone-Beam CT
  469. Online Resource Allocation with an Endogenous Markov State: Fewer LP Solves Earn More
  470. PEACE: Covariant learning of nonadiabatic manifolds with parity-resolved Hamiltonians
  471. How Do LLMs Change Predictions Under Negation?
  472. RELATE: An Evaluation Framework for measuring Relational Orientation of Large Language Models
  473. EvoSignal: LLM-Guided Evolutionary Design of Modular Traffic Signal Control Programs
  474. World Potential Model: Pretrained World Knowledge as Progress Potentials
  475. UniCSI Towards a Universal Wi-Fi CSI Encoder for Ubiquitous Human Sensing
  476. DrugTargetWorld: A Synthetic Biobank for Training and Benchmarking AI Scientists
  477. Reliability of LLM Judges for Evaluating Entity Alignment
  478. MARS: Malware Analysis with Rule-Based Scoring of LLM Claims
  479. Constitution-Guided Watermarking
  480. A Framework for the Systematic Review of ML Assets in AI Registries
  481. CircuitGate: Logic-Consistent Circuit-Level Functional Modeling for And-Inverter Graphs
  482. Certified by Abstention: Distribution-Free Guarantees for Chain-of-Thought Verifiers at Small Calibration Budgets
  483. WAPR: A Foundation Model for Wide-Angle Refinement in Unseen Object Pose Estimation
  484. KASALv2: Fully Automatic 3D Rotational Symmetry Classification and Axis Localization
  485. Unpaired Canonical Correlation Analysis
  486. A Comparative Study of Evaluation Metrics for Long-Document Financial Narrative Summarization with Transformers
  487. Align Before You Combine: Reference Space Calibration for Supervision Without Ground Truth
  488. Reflected Anchored Langevin Algorithms
  489. Differential Refresh Policies for Models Trained on Lagging Data Snapshots: From a Single-Age Equivalence Limit to an Optimal Per-Segment Allocation
  490. ActiveLang: Active Open-Vocabulary 3D Mapping with Semantic-Uncertainty-Guided Exploration
  491. OmniCam: Omni-Camera Trajectory Generation via Geometry-Grounded Pose Token Learning
  492. LiG-DETR: Local-in-Global Reassembly in Latent Space for Aerial Object Detection
  493. Physics-Informed Neural Plasticity: PDE Solvers That Reshape Themselves
  494. TERRA: Learning Transportable Latent Actions through Temporal Effect Representation and Relational Alignment
  495. Safe on Average, Unsafe in the Tail: When Is the Episodic-Cost Tail Controllable?
  496. Adjoint-Based Calibration and Optimal Control of Stochastic Multiscale Bioprocess Digital Twins
  497. RT-DETR-World: Transferring Rich LLM Semantics to Real-Time Open-Vocabulary Detection
  498. SpatialUQ: Post-Hoc Uncertainty Quantification from Spatial Consistency in Black-Box Vision Models
  499. Ream: Unfolding Mutual Awareness in Human-Agent Workspaces
  500. Cognitive Schemas, Laws and Tasks
  501. Gaussian Material Fields for Volumetric Multi-Energy CT Decomposition
  502. Correspondences as Decisions: JevNexus for Decision-Centric Schema Matching
  503. Mitigating Accent-Language Confusion in Self-Supervised Speech Representations for Language Identification
  504. InstanceBench: Diagnosing Referential Reasoning and Target Identity in Referring Expression Segmentation
  505. CHASE: Channel-Aligned Structure Exploitation for Geometry-Aware Model Engineering
  506. MORA: Modeling Observed Changes for Drift-Robust Time-Series Anomaly Detection
  507. When Should an In-Context Learner Expand Its Hypothesis Space?
  508. Boundary-Free Contextual Biasing: Depth-Adaptive Gating and Reading-Space Matching for Unsegmented Languages
  509. BanglaRhet: Benchmarking Classical and Transformer Models for Rhetorical and Persuasion Detection in Bangla Political Speech
  510. An Invariant Tangent-Angle Descriptor and a Band U-Net for 2D Fragment Adjacency Prediction
  511. Right Number, Wrong State? Measuring Cross-Jurisdiction Substitution in LLM Recall of State Policy
  512. DSReg: Provably Recovering Individual World Latents without Reconstruction
  513. RLHND: Video Foundation Models as Physically Grounded Hand Trackers for Robot Learning
  514. Finite-Rank Logistic Gaussian Processes with Exact Likelihood for Conditional Density Estimation
  515. From Global Alignment to Local Grounding: Zero-Shot Chinese Character Recognition with Radical Verification
  516. Arctic Questions, Missing Answers: A Dataset and Benchmark for LLM Abstention in Arctic Science
  517. LighTROcc: Lightweight 4D Occupancy Forecasting via Instance-Centric 3D Gaussians
  518. TIRA: Tumor Immune Representation Adaptation for Zero-Shot Cross-Cancer MSI and TMB Prediction
  519. InscriptionOCR: A Dataset and Method for Understanding Inscriptions
  520. An extended deep energy method for thermo-mechanical crack propagation
  521. SkillCycle: Co-Evolving Agent Policies and Skill Banks
  522. RSI-Forge: From Research Papers to Environments for Recursive Self-Improvement
  523. What a Reporting Convention Hides: A Matched-Budget Audit of Quantum Natural Gradient with an Exactly Computed Metric
  524. Spatial Latent Reasoning for Embodied Reference Understanding
  525. trACT: temporal revelation Airborne Camera Trap
  526. Self-Consuming Generative Models with Co-Evolving Human Preferences
  527. Shared Geometry As A Rosetta Stone: Cross-Modal Alignment Without Paired Data
  528. Stability and Diversity of Networked Self-Consuming Generative Ecosystems
  529. TiTok: Audio-Visual LLM for Multi-Segment Temporal Grounding
  530. Let the Library Speak: Self-Advertised Method Selection for Formal Proving
  531. TutorLoop: Regulating Student Learning Behaviors via Sensor-in-the-Loop Generative Feedback
  532. ARCS: Towards Precise Text-to-SQL via Structured Disambiguation
  533. Quantifying Volumetric Risk: Class-Aware Asymmetric Weighted Conformal Prediction for 3D Medical Image Segmentation
  534. The Persona Hierarchy Model: Understanding Contextual Generalization in Fine-Tuning LLMs
  535. PPCAR-Net: Projection-Refined Parametric 3D Coronary Artery Reconstruction from Sparse X-ray Angiographic Views
  536. ScribbleEdit: A Benchmark for Scribble-Only Image Editing
  537. The Confidence Game: Strategic Miscalibration in Human-AI Delegation
  538. Closing the Loop on Contrail Avoidance with Satellite Verification
  539. TopoGraphRAG-Bench: Evaluating Multimodal GraphRAG on Layout-Grounded Evidence Reasoning
  540. Global Exponential Convergence of Two-Layer Linear Network Training
  541. OnlineQAT: On-Policy Distillation for Ultra-Low-Bit Large Language Models
  542. Do Image Editors Follow Depth-Dependent Blur and Aperture Response? A Rendered-Ground-Truth Pilot Audit
  543. TileSkipper: Region-Adaptive Tile Pruning for 3D Gaussian Splatting
  544. Benign Overfitting under Heterogeneous Input Fusion
  545. Ask the Expert: LLM-Guided Reinforcement Learning for Autonomous Cyber Defense
  546. VM-ARRAYDPS: Virtual Microphone Augmented Diffusion Posterior Sampling for Unsupervised Blind Speech Separation
  547. CRT-HMAR: Causal Requirement Tracing-Guided Hierarchical Multi-Agent Regulation for Open-Task-Aware Infrared-Visible Image Fusion
  548. GRC-Net: Global Representation Consistency Network for Unsupervised Multimodal Anomaly Detection
  549. Visual Jev Rewards: Reference-Bound Verification for Multi-Subject Image Generation
  550. VIS-Ground: Video Interactive Storytelling with Contextual Grounding
  551. Dialect-Robust Speech Language Models with Synthetic Pseudo-Dialect Augmentation
  552. SAREO-FM: Decoupled Semantic Supervision for SAR-EO Foundation Models
  553. The Geometry of Anisotropic Dilation for Optimal Regularization
  554. MovieSTAGE: Scene, Transition, and Global Encoding for Movie-fMRI ADHD Classification
  555. emg2face: Expressive Facial Animation with High-Density Surface EMG
  556. Node-level Graph Neural Architecture Search Framework
  557. The AI Evaluation Ecosystem
  558. FaceKit: a Toolkit for Interpretable Facial Phenotyping, Synthetic Image Generation and Privacy Analysis in Rare Diseases
  559. Self-attention summary networks for subsurface velocity-model building from common-image gathers
  560. Twist Flow for Inverse Problems
  561. CATune: Structural Constraint-Aware Bayesian Optimization for DBMS Configuration Tuning
  562. EM-SNN: Efficiently Modulated Spiking Neural Network for Remote Sensing Image Dehazing
  563. Hardware-aware Calibrated Clustered Attention for Efficient Visual Geometric Transformers
  564. Evaluating Trajectory Features for Routing Final-Layer Attention
  565. The Symbol of the Surrogate: Measuring Numerical Provenance in Neural PDE Solvers
  566. PhyDiCT: Plug-and-Play CT Reconstruction from Sparse X-Rays via Differentiable Rendering and Strong Priors
  567. An Informational Curse of Horizon in Goal-Conditioned Policy Learning
  568. Beyond Nominal Equilibria: Risk-Averse Multi-Population Mean-Field Games
  569. Pooling Representation Autoencoders for Efficient Diffusion