mlsanityData to 14 Oct 2026

Papers

New on arXiv

Thu 7 May 2026

811 papers

  1. It Takes 8 Tokens: Weak-to-Strong Off-Policy RL via Auxiliary BranchesLLM reasoning → Reinforcement learning for reasoning
  2. Masked Diffusion Language Models are Strong and Steerable Text-Based World Models for Agentic RLWorld models → Latent dynamics models and model-based RL +2
  3. ReasonLab: A Controlled and Auditable Evaluation of Prompting Techniques for Multiple-Choice QALLM reasoning → Prompting and chain-of-thought reasoning
  4. Eta Given Delta: Defining LLM Tool Efficiency With Marginal Tool UtilityLLM agents and tool use → Tool learning and function calling
  5. Polestar: Drift-Aware Cache Calibration and Token Commitment for Efficient Inference of Diffusion LLMsEfficient LLM inference and serving → Diffusion LLM and MoE inference
  6. Token Time Continuous Diffusion for Language ModelingDiffusion models → Discrete and text diffusion
  7. G-SHARE: A Guideline-Based Structured Reasoning Framework for Human-Factor Event DiagnosisLLM reasoning → Clinical and biomedical applications
  8. ADEPT: An Entropy-Driven Dual-Strategy Agent for Interactive Video RetrievalLLM agents and tool use → Multimodal and embodied agents
  9. Formalizing Latent Thoughts: Four Axioms of Thought Representation in LLMsLLM reasoning → Test-time compute and efficient thinking
  10. Towards Distributed Inference of LLMs on a P2P NetworkEfficient LLM inference and serving → Serving systems and request scheduling
  11. A Simple State Space Model Excels at Multivariate Time Series ClassificationState-space sequence models (S4, Mamba) → Core SSM architectures for language
  12. Stop Comparing LLM Agents Without Disclosing the HarnessLLM agents and tool use → Software engineering coding agents
  13. Preference Instability in Reward Models: Detection and Mitigation via Sparse AutoencodersLLM alignment / RLHF / preference optimisation → Reward modeling and RLHF +1
  14. On Seeding Watermarks to Detect Verbatim LLM Copy-Paste ResponsesLLM safety and security → Watermarking generated content and models
  15. Computer Use at the Edge of the Statistical PrecipiceLLM agents and tool use → GUI and mobile agents
  16. Designing Intelligent Enterprise Agents: A Capability-Aligned Multi-Agent ArchitectureLLM agents and tool use → General LLM agent frameworks and benchmarks
  17. Research on Security Enhancement Methods for Adversarial Robust Large Language Model Intelligent Agents for Medical Decision-Making TasksLLM agents and tool use → Medical and scientific domain agents
  18. HyperTransport: Amortized Conditioning of T2I Generative ModelsMechanistic interpretability → Activation steering methods
  19. Path-Coupled Bellman Flows for Distributional Reinforcement LearningOffline RL and imitation learning from data → Diffusion and flow policies for offline RL +1
  20. Multimodal Emotion Recognition via Causal-Diffusion Bridge (Affect-Diff)Diffusion models → Audio, speech and music
  21. Why Do DiT Editors Drift? Plug-and-Play Low Frequency Alignment in VAE Latent SpaceDiffusion models → Image generation, text-to-image and editing
  22. When Language Overwrites Vision: Over-Alignment and Geometric Debiasing in Vision-Language ModelsMultimodal LLMs / vision-language models → VLM benchmarks and visual reasoning
  23. TinySSL: Distilled Self-Supervised Pretraining for Sub-Megabyte MCU ModelsKnowledge distillation → General knowledge distillation methods
  24. When Does Value-Aware KV Eviction Help? A Fixed-Contract Diagnostic for Non-Monotone Cache CompressionEfficient LLM inference and serving → KV cache optimization
  25. Towards Differentially Private Reinforcement Learning with General Function ApproximationDeep reinforcement learning methods → RL for energy, networks and operations
  26. The Context Gathering Decision Process: A POMDP Framework for Agentic SearchLLM agents and tool use → Agent memory systems
  27. Cognitive Agent Compilation for Explicit Problem Solver ModelingLLM agents and tool use → Educational agents and math problem solving
  28. PACEvolve++: Improving Test-time Learning for Evolutionary Search AgentsLLM agents and tool use → General LLM agent frameworks and benchmarks
  29. A Systematic Investigation of RL-Jailbreaking in LLMsLLM safety and security → Jailbreaking
  30. Behavior Cue Reasoning: Monitorable Reasoning Improves Efficiency and Safety through OversightLLM reasoning → Test-time compute and efficient thinking
  31. LensVLM: Selective Context Expansion for Compressed Visual Representation of TextMultimodal LLMs / vision-language models → Visual token pruning and efficiency
  32. CoBit: Language Modeling with Bitstream DiffusionDiffusion models → Discrete and text diffusion
  33. Dual-Agent Co-Training for Health Coaching via Implicit Adversarial Preference OptimizationLLM agents and tool use → Medical and scientific domain agents +1
  34. SmellBench: Evaluating LLM Agents on Architectural Code Smell RepairLLM agents and tool use → Software engineering coding agents
  35. Echo: KV-Cache-Free Associative Recall with Spectral Koopman OperatorsLLM reasoning → Test-time compute and efficient thinking +2
  36. Why Does Agentic Safety Fail to Generalize Across Tasks?LLM agents and tool use → Agent threats, guardrails and governance
  37. Response Time Enhances Alignment with Heterogeneous PreferencesLLM alignment / RLHF / preference optimisation → LLM preference optimization methods
  38. PLOT: Progressive Localization via Optimal Transport in Neural Causal AbstractionMechanistic interpretability → Circuits and mechanistic analysis
  39. f-Divergence Regularized RLHF: Two Tales of Sampling and Unified AnalysesLLM alignment / RLHF / preference optimisation → Reward modeling and RLHF
  40. Bringing Multimodal Large Language Models to Infrared-Visible Image Fusion Quality AssessmentMultimodal LLMs / vision-language models → MLLM training and visual instruction tuning
  41. From Surface Learning to Deep Understanding: A Grounded AI Tutoring System for MoodleRetrieval-augmented generation → Agentic RAG, agents and memory
  42. Learning and Reusing Policy Decompositions for Hierarchical Generalized Planning with LLM AgentsLLM agents and tool use → Reinforcement learning for LLM agents
  43. Multi-Objective Constraint Inference using Inverse reinforcement learningDeep reinforcement learning methods → RL for games and combinatorial optimization
  44. Bridging the Last Mile of Circuit Design: PostEDA-Bench, a Hierarchical Benchmark for PPA Convergence and DRC FixingLLM agents and tool use → Software engineering coding agents
  45. Physics-Based Flow Matching for Full-Field Prediction of Silicon Photonic DevicesDiffusion models → Flow matching, time series and science
  46. A2RD: Agentic Autoregressive Diffusion for Long Video ConsistencyLLM agents and tool use → Multimodal and embodied agents
  47. Can LLMs Take Retrieved Information with a Grain of Salt?Retrieval-augmented generation → RAG robustness and adaptive retrieval
  48. Regulating Branch Parallelism in LLM ServingEfficient LLM inference and serving → Serving systems and request scheduling
  49. Same Signal, Opposite Meaning: Direction-Informed Adaptive Learning for LLM AgentsLLM reasoning → Reinforcement learning for reasoning +1
  50. TraXion: Rethinking Pre-training Frameworks for Mobility and BeyondLLM reasoning → Spatial and embodied reasoning
  51. Self-Programmed Execution for Language-Model AgentsLLM agents and tool use → Software engineering coding agents
  52. MIST: Multimodal Interactive Speech-based Tool-calling Conversational Assistants for Smart HomesLLM agents and tool use → General LLM agent frameworks and benchmarks
  53. Mitigating Cognitive Bias in RLHF by Altering RationalityLLM alignment / RLHF / preference optimisation → Reward modeling and RLHF
  54. Not All Tokens Need 40 Steps: Heterogeneous Step Allocation in Diffusion Transformers for Efficient Video GenerationDiffusion models → Video generation
  55. Beyond the Black Box: Interpretability of Agentic AI Tool UseLLM agents and tool use → Software engineering coding agents +1
  56. Don't Retrain, Align: Adapting Autoregressive LMs to Diffusion LMs via Representation AlignmentDiffusion models → Discrete and text diffusion
  57. How Well Do LLMs Perform on the Simplest Long-Chain Reasoning Tasks: An Empirical Study on the Equivalence Class ProblemLLM reasoning → Reasoning benchmarks and evaluation
  58. On the Divergence of Differential Temporal Difference Learning without Local ClocksDeep reinforcement learning methods → RL theory and sample complexity
  59. Agentick: A Unified Benchmark for General Sequential Decision-Making AgentsLLM agents and tool use → General LLM agent frameworks and benchmarks
  60. Christoffel-DPS: Optimal sensor placement in diffusion posterior sampling for arbitrary distributionsDiffusion models → Inverse problems and image restoration
  61. How to Compress KV Cache in RL Post-Training? Shadow Mask Distillation for Memory-Efficient AlignmentLLM reasoning → Reinforcement learning for reasoning +2
  62. Narrow Secret Loyalty Dodges Black-Box AuditsLLM safety and security → Membership inference attacks
  63. AGWM: Affordance-Grounded World Models for Environments with Compositional PrerequisitesWorld models → World models for language agents
  64. Extracting Search Trees from LLM Reasoning Traces Reveals Myopic PlanningLLM reasoning → Reasoning for agents, planning and code
  65. Why DDIM Hallucinates More Than DDPM: A Theoretical Analysis of Reverse DynamicsDiffusion models → Medical imaging and segmentation
  66. A Unified Measure-Theoretic View of Diffusion, Score-Based, and Flow Matching Generative ModelsDiffusion models → Diffusion foundations and sampling
  67. Randomness is sometimes necessary for coordinationDeep reinforcement learning methods → RL for games and combinatorial optimization
  68. SHARP: A Self-Evolving Human-Auditable Rubric Policy for Financial Trading AgentsLLM agents and tool use → Financial trading and forecasting agents +1
  69. A Theory of Online Learning with Autoregressive Chain-of-Thought ReasoningLLM reasoning → Prompting and chain-of-thought reasoning
  70. Towards Security-Auditable LLM Agents: A Unified Graph RepresentationLLM agents and tool use → Agent threats, guardrails and governance
  71. Conformal Agent Error AttributionLLM agents and tool use → Software engineering coding agents
  72. Distributional Process Reward Models: Calibrated Prediction of Future Rewards via Conditional Optimal TransportLLM reasoning → Reinforcement learning for reasoning +1
  73. When Does Critique Improve AI-Assisted Theoretical Physics? SCALAR: Structured Critic--Actor Loop for Agentic ReasoningLLM reasoning → Reasoning for agents, planning and code +1
  74. Revisiting Adam for Streaming Reinforcement LearningDeep reinforcement learning methods → Deep RL algorithms and representations
  75. Sparse Attention as a Range Searching Problem: Towards an Inference-Efficient Index for KV CacheEfficient LLM inference and serving → KV cache optimization
  76. Weblica: Scalable and Reproducible Training Environments for Visual Web AgentsLLM agents and tool use → Web, search, and research agents
  77. R3L: Reasoning 3D Layouts from Relative Spatial RelationsLLM reasoning → Spatial and embodied reasoning
  78. Gradient Extrapolation-Based Policy OptimizationLLM reasoning → Reinforcement learning for reasoning
  79. A Self-Healing Framework for Reliable LLM-Based Autonomous AgentsLLM agents and tool use → Software engineering coding agents
  80. On Training in ImaginationLLM alignment / RLHF / preference optimisation → Reward modeling and RLHF +1
  81. When Routine Chats Turn Toxic: Unintended Long-Term State Poisoning in Personalized AgentsLLM agents and tool use → Agent threats, guardrails and governance
  82. OmicsLM: A Multimodal Large Language Model for Multi-Sample Omics ReasoningLLM reasoning → Clinical and biomedical applications +1
  83. When Does a Language Model Commit? A Finite-Answer Theory of Pre-Verbalization CommitmentLLM reasoning → Test-time compute and efficient thinking
  84. Conditional generation of antibody sequences with classifier-guided germline-absorbing discrete diffusionDiffusion models → Flow matching, time series and science
  85. Agentic Coding Needs Proactivity, Not Just AutonomyLLM agents and tool use → Software engineering coding agents
  86. From Storage to Experience: A Survey on the Evolution of LLM Agent Memory MechanismsLLM agents and tool use → Agent memory systems
  87. UniPool: A Globally Shared Expert Pool for Mixture-of-ExpertsMixture-of-experts → Expert routing strategies
  88. EMO: Pretraining Mixture of Experts for Emergent ModularityMixture-of-experts → MoE large language models
  89. Verifier-Backed Hard Problem Generation for Mathematical ReasoningLLM reasoning → Mathematical reasoning
  90. Relit-LiVE: Relight Video by Jointly Learning Environment VideoDiffusion models → 3D, point cloud and scene generation
  91. When No Benchmark Exists: Validating Comparative LLM Safety Scoring Without Ground-Truth LabelsLLM safety and security → Safety alignment, refusal and guardrails
  92. AI co-mathematician: Accelerating mathematicians with agentic AILLM agents and tool use → Educational agents and math problem solving
  93. Beyond Negative Rollouts: Positive-Only Policy Optimization with Implicit Negative GradientsLLM reasoning → Reinforcement learning for reasoning +1
  94. Superintelligent Retrieval Agent: The Next Frontier of Agentic RetrievalRetrieval-augmented generation → Agentic RAG, agents and memory +1
  95. StraTA: Incentivizing Agentic Reinforcement Learning with Strategic Trajectory AbstractionLLM agents and tool use → Reinforcement learning for LLM agents
  96. Recursive Agent OptimizationDeep reinforcement learning methods → RL training of language and agent models
  97. Can RL Teach Long-Horizon Reasoning to LLMs? Expressiveness Is KeyLLM reasoning → Reinforcement learning for reasoning
  98. Cited but Not Verified: Parsing and Evaluating Source Attribution in LLM Deep Research AgentsLLM agents and tool use → Web, search, and research agents
  99. MASPO: Joint Prompt Optimization for LLM-based Multi-Agent SystemsLLM agents and tool use → General LLM agent frameworks and benchmarks
  100. SkillOS: Learning Skill Curation for Self-Evolving AgentsLLM agents and tool use → Agent skill learning and evolution +1
All 811
  1. SoftSAE: Dynamic Top-K Selection for Adaptive Sparse AutoencodersMechanistic interpretability → Sparse autoencoder training and evaluation
  2. AI CFD Scientist: Toward Open-Ended Computational Fluid Dynamics Discovery with Physics-Aware AI AgentsLLM agents and tool use → Web, search, and research agents
  3. How Many Iterations to Jailbreak? Dynamic Budget Allocation for Multi-Turn LLM EvaluationLLM safety and security → Jailbreaking
  4. Patch2Vuln: Agentic Reconstruction of Vulnerabilities from Linux Distribution Binary PatchesLLM agents and tool use → Agent threats, guardrails and governance
  5. UniSD: Towards a Unified Self-Distillation Framework for Large Language ModelsKnowledge distillation → Knowledge distillation for LLMs
  6. FedAttr: Towards Privacy-preserving Client-Level Attribution in Federated LLM Fine-tuningLLM safety and security → Federated unlearning and poisoning
  7. ReActor: Reinforcement Learning for Physics-Aware Motion RetargetingOffline RL and imitation learning from data → Real-world robot offline RL +1
  8. DINORANKCLIP: DINOv3 Distillation and Injection for Vision-Language Pretraining with High-Order Ranking ConsistencyMultimodal LLMs / vision-language models → Prompt learning and adaptation of CLIP
  9. NeuroAgent: LLM Agents for Multimodal Neuroimaging Analysis and ResearchLLM agents and tool use → Medical and scientific domain agents
  10. Improved techniques for fine-tuning flow models via adjoint matching: a deterministic control pipelineLLM alignment / RLHF / preference optimisation → Diffusion and multimodal model alignment
  11. Diverse Sampling in Diffusion Models with Marginal Preserving Particle GuidanceDiffusion models → Diffusion foundations and sampling
  12. Sequential Design of Genetic Circuits Under Uncertainty With Reinforcement LearningDeep reinforcement learning methods → RL for energy, networks and operations
  13. Continuous Latent Diffusion Language ModelDiffusion models → Discrete and text diffusion
  14. Diffusion-Based Posterior Sampling: A Feynman-Kac Analysis of Bias and StabilityDiffusion models → Inverse problems and image restoration
  15. MedHorizon: Towards Long-context Medical Video Understanding in the WildMultimodal LLMs / vision-language models → Medical vision-language models
  16. STALE: Can LLM Agents Know When Their Memories Are No Longer Valid?LLM agents and tool use → Agent memory systems
  17. On the Implicit Reward Overfitting and the Low-rank Dynamics in RLVRLLM reasoning → Reinforcement learning for reasoning
  18. Agentic AIs Are the Missing Paradigm for Out-of-Distribution Generalization in Foundation ModelsLLM agents and tool use → General LLM agent frameworks and benchmarks
  19. Learning to Cut: Reinforcement Learning for Benders DecompositionDeep reinforcement learning methods → RL for games and combinatorial optimization
  20. DCR: Counterfactual Attractor Guidance for Rare Compositional GenerationDiffusion models → Diffusion foundations and sampling
  21. FreeSpec: Training-Free Long Video Generation via Singular-Spectrum ReconstructionDiffusion models → Video generation
  22. MARBLE: Multi-Aspect Reward Balance for Diffusion RLDiffusion models → Diffusion and flow policies for robotics +2
  23. Operator-Guided Invariance Learning for Continuous Reinforcement LearningDeep reinforcement learning methods → RL theory and sample complexity
  24. From Token Lists to Graph Motifs: Weisfeiler-Lehman Analysis of Sparse Autoencoder FeaturesMechanistic interpretability → SAE features and their applications
  25. Instrumental Choices: Measuring the Propensity of LLM Agents to Pursue Instrumental BehaviorsLLM agents and tool use → General LLM agent frameworks and benchmarks
  26. Autonomous Adversary: Red-Teaming in the age of LLMLLM agents and tool use → Agent threats, guardrails and governance +1
  27. ReasonSTL: Bridging Natural Language and Signal Temporal Logic via Tool-Augmented Process-Rewarded LearningLLM reasoning → Spatial and embodied reasoning
  28. OA-WAM: Object-Addressable World Action Model for Robust Robot ManipulationWorld models → Embodied and robot world models +1
  29. Patch-Effect Graph Kernels for LLM InterpretabilityMechanistic interpretability → Circuits and mechanistic analysis
  30. GeoStack: A Framework for Quasi-Abelian Knowledge Composition in VLMsMultimodal LLMs / vision-language models → Embodied and 3D vision-language models
  31. Efficient Serving for Dynamic Agent Workflows with Prediction-based KV-Cache ManagementLLM agents and tool use → Software engineering coding agents +1
  32. Hitting Time Isomorphism for Multi-Stage Planning with Foundation PoliciesOffline RL and imitation learning from data → Offline RL algorithms +1
  33. To What Extent Does Agent-generated Code Require Maintenance? An Empirical StudyLLM agents and tool use → Software engineering coding agents
  34. Beyond Task Success: Measuring Workflow Fidelity in LLM-Based Agentic Payment SystemsLLM agents and tool use → Software engineering coding agents
  35. PrefixGuard: From LLM-Agent Traces to Online Failure-Warning MonitorsLLM agents and tool use → Software engineering coding agents
  36. Scene-Adaptive Continual Learning for CSI-based Human Activity Recognition with Mixture of ExpertsMixture-of-experts → Mixture of LoRA experts
  37. Constraint Decay: The Fragility of LLM Agents in Backend Code GenerationLLM agents and tool use → Software engineering coding agents
  38. SCRuB: Social Concept Reasoning under Rubric-Based EvaluationLLM reasoning → Reasoning benchmarks and evaluation
  39. AgenticPrecoding: LLM-Empowered Multi-Agent System for Precoding OptimizationLLM agents and tool use → Agentic AI for wireless networks
  40. Knowledge Graphs, the Missing Link in Agentic AI-based Formal VerificationLLM agents and tool use → Software engineering coding agents
  41. Pop Quiz Attack: Black-box Membership Inference Attacks Against Large Language ModelsLLM safety and security → Membership inference attacks
  42. FREPix: Frequency-Heterogeneous Flow Matching for Pixel-Space Image GenerationDiffusion models → Flow matching, time series and science
  43. E = T*H/(O+B): A Dimensionless Control Parameter for Mixture-of-Experts EcologyMixture-of-experts → MoE theory and gating functions
  44. Constraining Host-Level Abuse in Self-Hosted Computer-Use Agents via TEE-Backed IsolationLLM agents and tool use → Agent threats, guardrails and governance
  45. Reconstruction or Semantics? What Makes a Latent Space Useful for Robotic World ModelsWorld models → Embodied and robot world models +1
  46. Asymmetric On-Policy Distillation: Bridging Exploitation and Imitation at the Token LevelKnowledge distillation → On-policy and self-distillation for LLMs +1
  47. Independent Learning of Nash Equilibria in Partially Observable Markov Potential Games with Decoupled DynamicsDeep reinforcement learning methods → RL for games and combinatorial optimization
  48. Continuous-Time Distribution Matching for Few-Step Diffusion DistillationDiffusion models → Diffusion foundations and sampling
  49. A Unified Pair-GRPO Family: From Implicit to Explicit Preference Constraints for Stable and General RL AlignmentLLM reasoning → Reinforcement learning for reasoning +1
  50. Beyond the Independence Assumption: Finite-Sample Guarantees for Deep Q-Learning under τ-MixingDeep reinforcement learning methods → Deep RL algorithms and representations
  51. The Interplay of Data Structure and Imbalance in the Learning Dynamics of Diffusion ModelsDiffusion models → Diffusion foundations and sampling
  52. Layer Collapse in Diffusion Language ModelsDiffusion models → Discrete and text diffusion
  53. From Agent Loops to Deterministic Graphs: Execution Lineage for Reproducible AI-Native WorkLLM agents and tool use → General LLM agent frameworks and benchmarks
  54. Flow Matching with Arbitrary Auxiliary PathsDiffusion models → Flow matching, time series and science
  55. More Than Can Be Said: A Benchmark and Framework for Pre-Question Scientific IdeationLLM agents and tool use → Web, search, and research agents
  56. Don't Lose Focus: Activation Steering via Key-Orthogonal ProjectionsMechanistic interpretability → Activation steering methods
  57. A Regime Theory of Controller Class Selection for LLM Action DecisionsMultimodal LLMs / vision-language models → Embodied and 3D vision-language models
  58. Earth-o1: A Grid-free Observation-native Atmospheric World ModelWorld models → Latent predictive world models
  59. MANTRA: Synthesizing SMT-Validated Compliance Benchmarks for Tool-Using LLM AgentsLLM agents and tool use → Tool learning and function calling
  60. Teaching Thinking Models to Reason with Tools: A Full-Pipeline Recipe for Tool-Integrated ReasoningLLM reasoning → Reasoning for agents, planning and code +1
  61. Improving the Efficiency of Language Agent Teams with Adaptive Task GraphsLLM agents and tool use → General LLM agent frameworks and benchmarks
  62. Measuring Black-Box Confidence via Reasoning Trajectories: Geometry, Coverage, and VerbalizationLLM reasoning → Test-time compute and efficient thinking
  63. Molecules Meet Language: Confound-Aware Representation Learning and Chemical Property Steering in Transformer-VAE Latent SpacesMechanistic interpretability → Interpreting biomedical and science models
  64. Render, Don't Decode: Weight-Space World Models with Latent Structural DisentanglementWorld models → Video-generation world models and simulators
  65. LatentRAG: Latent Reasoning and Retrieval for Efficient Agentic RAGLLM reasoning → Knowledge- and tool-augmented reasoning +2
  66. Eulerian Motion Guidance: Robust Image Animation via Bidirectional Geometric ConsistencyDiffusion models → Video generation
  67. A Flow Matching Algorithm for Many-Shot Adaptation to Unseen DistributionsDiffusion models → Flow matching, time series and science
  68. Spark3R: Asymmetric Token Reduction Makes Fast Feed-Forward 3D ReconstructionEfficient LLM inference and serving → Token pruning and merging for ViTs
  69. Can Attribution Predict Risk? From Multi-View Attribution to Planning Risk Signals in End-to-End Autonomous DrivingEnd-to-end autonomous driving → End-to-end system architecture
  70. CKT-WAM: Parameter-Efficient Context Knowledge Transfer Between World Action ModelsWorld models → Embodied and robot world models
  71. Rethinking RL for LLM Reasoning: It's Sparse Policy Selection, Not Capability LearningLLM reasoning → Reinforcement learning for reasoning
  72. Profiling for Pennies: Unveiling the Privacy Iceberg of LLM AgentsLLM agents and tool use → Agent threats, guardrails and governance
  73. Safactory: A Scalable Agentic Infrastructure for Training Trustworthy Autonomous IntelligenceLLM agents and tool use → General LLM agent frameworks and benchmarks
  74. Soft Deterministic Policy Gradient with Gaussian SmoothingDeep reinforcement learning methods → RL theory and sample complexity
  75. Memory Inception: Latent-Space KV Cache Manipulation for Steering LLMsEfficient LLM inference and serving → KV cache optimization +1
  76. When to Trust Imagination: Adaptive Action Execution for World Action ModelsWorld models → Embodied and robot world models
  77. UniPrefill: Universal Long-Context Prefill Acceleration via Block-wise Dynamic SparsificationEfficient LLM inference and serving → KV cache optimization
  78. Joint Consistency: A Unified Test-Time Aggregation Framework via Energy MinimizationLLM reasoning → Test-time compute and efficient thinking
  79. Beyond Fixed Benchmarks and Worst-Case Attacks: Dynamic Boundary Evaluation for Language ModelsLLM safety and security → Safety alignment, refusal and guardrails
  80. Playing the network backward: A Game Theoretic Attribution FrameworkMechanistic interpretability → Circuits and mechanistic analysis
  81. Federation of Experts: Communication Efficient Distributed Inference for Large Language ModelsMixture-of-experts → Efficient MoE training and inference systems
  82. A2TGPO: Agentic Turn-Group Policy Optimization with Adaptive Turn-level ClippingLLM agents and tool use → Reinforcement learning for LLM agents
  83. EA-WM: Event-Aware Generative World Model with Structured Kinematic-to-Visual Action FieldsWorld models → Embodied and robot world models +1
  84. OPSD Compresses What RLVR Teaches: A Post-RL Compaction Stage for Reasoning ModelsLLM reasoning → Prompting and chain-of-thought reasoning +1
  85. Event-Causal RAG: A Retrieval-Augmented Generation Framework for Long Video Reasoning in Complex ScenariosLLM reasoning → Multimodal reasoning +2
  86. BioMedArena: An Open-source Toolkit for Building and Evaluating Biomedical Deep Research AgentsLLM agents and tool use → Medical and scientific domain agents
  87. VLA-GSE: Boosting Parameter-Efficient Fine-Tuning in VLA with Generalized and Specialized ExpertsVision-language-action models → Generalist VLA models and benchmarks
  88. Retina-RAG: Retrieval-Augmented Vision-Language Modeling for Joint Retinal Diagnosis and Clinical Report GenerationMultimodal LLMs / vision-language models → Medical vision-language models
  89. Mean Mode Screaming: Mean--Variance Split Residuals for 1000-Layer Diffusion TransformersDiffusion models → Diffusion foundations and sampling
  90. Post-Reasoning: Improving the Performance of Non-Thinking Models at No CostLLM reasoning → Test-time compute and efficient thinking
  91. Beyond Accuracy: Policy Invariance as a Reliability Test for LLM Safety JudgesLLM safety and security → Safety alignment, refusal and guardrails
  92. Stateful Agent BackdoorLLM agents and tool use → Agent threats, guardrails and governance +1
  93. Entropy-Regularized Adjoint Matching for Offline Reinforcement LearningDiffusion models → Diffusion and flow policies for robotics +1
  94. Secure Seed-Based Multi-bit Watermarking for Diffusion Models from First PrinciplesDiffusion models → Robustness, privacy and adversarial aspects +1
  95. AdaGamma: State-Dependent Discounting for Temporal Adaptation in Reinforcement LearningDeep reinforcement learning methods → RL theory and sample complexity
  96. Unifying Goal-Conditioned RL and Unsupervised Skill Learning via Control-MaximizationDeep reinforcement learning methods → Deep RL algorithms and representations
  97. Listwise Policy Optimization: Group-based RLVR as Target-Projection on the LLM Response SimplexLLM reasoning → Reinforcement learning for reasoning
  98. MemReranker: Reasoning-Aware Reranking for Agent Memory RetrievalLLM reasoning → Reasoning for agents, planning and code +1
  99. Skill1: Unified Evolution of Skill-Augmented Agents via Reinforcement LearningLLM agents and tool use → Agent skill learning and evolution +1
  100. Breaking, Stale, or Missing? Benchmarking Coding Agents on Project-Level Test EvolutionLLM agents and tool use → Software engineering coding agents
  101. P-Guide: Parameter-Efficient Prior Steering for Single-Pass CFG InferenceDiffusion models → Flow matching, time series and science
  102. Pest-Thinker: Learning to Think and Reason like Entomologists via Reinforcement LearningLLM reasoning → Clinical and biomedical applications +1
  103. Policy-Guided Stepwise Model Routing for Cost-Effective ReasoningLLM reasoning → Test-time compute and efficient thinking
  104. CrossCult-KIBench: A Benchmark for Cross-Cultural Knowledge Insertion in MLLMsMultimodal LLMs / vision-language models → VLM benchmarks and visual reasoning
  105. Tackling the Data-Parallel Load Balancing Bottleneck in LLM Serving: Practical Online Routing at ScaleEfficient LLM inference and serving → Serving systems and request scheduling
  106. Dynamic Pondering Sparsity-aware Mixture-of-Experts Transformer for Event Stream based Visual Object TrackingMixture-of-experts → Vision mixture-of-experts
  107. On Time, Within Budget: Constraint-Driven Online Resource Allocation for Agentic WorkflowsLLM agents and tool use → Software engineering coding agents
  108. Beyond Autoregressive RTG: Conditioning via Injection Outside Sequential Modeling in Decision TransformerOffline RL and imitation learning from data → Decision transformers and sequence modeling +1
  109. Uncovering Entity Identity Confusion in Multimodal Knowledge EditingMultimodal LLMs / vision-language models → Image editing and quality assessment
  110. VISD: Enhancing Video Reasoning via Structured Self-DistillationLLM reasoning → Multimodal reasoning +1
  111. Milestone-Guided Policy Learning for Long-Horizon Language AgentsLLM agents and tool use → Reinforcement learning for LLM agents
  112. Understanding diffusion models requires rethinking (again) generalizationDiffusion models → Diffusion foundations and sampling
  113. Navigating by Old Maps: The Pitfalls of Static Mechanistic Localization in LLM Post-TrainingMechanistic interpretability → Circuits and mechanistic analysis
  114. Arena as Offline Reward: Efficient Fine-Grained Preference Optimization for Diffusion ModelsDiffusion models → Diffusion and flow policies for robotics +1
  115. VibeServe: Can AI Agents Build Bespoke LLM Serving Systems?LLM agents and tool use → Software engineering coding agents
  116. Causal Reinforcement Learning for Complex Card Games: A Magic The Gathering BenchmarkWorld models → Latent dynamics models and model-based RL +1
  117. Towards Self-Explainable Document Visual Question Answering with Chain-of-Explanation PredictionsMultimodal LLMs / vision-language models → VLM benchmarks and visual reasoning
  118. Relay Buffer Independent Communication over Pooled HBM for Efficient MoE Inference on AscendMixture-of-experts → Efficient MoE training and inference systems
  119. Fusion in Your Way: Aligning Image Fusion with Heterogeneous Demands via Direct Preference OptimizationDiffusion models → Image generation, text-to-image and editing
  120. Requests of a Feather Must Flock Together: Batch Size vs. Prefix Homogeneity in LLM InferenceEfficient LLM inference and serving → KV cache optimization
  121. Novelty-based Tree-of-Thought Search for LLM Reasoning and PlanningLLM reasoning → Test-time compute and efficient thinking
  122. Optimal Transport for LLM Reward Modeling from Noisy PreferenceLLM alignment / RLHF / preference optimisation → Reward modeling and RLHF
  123. Does Synthetic Data Help? Empirical Evidence from Deep Learning Time Series ForecastersSynthetic data and self-improvement for LLMs → Model collapse from synthetic data
  124. Multi-agent decision making: A Blackwell's informativeness approachLLM agents and tool use → Social simulation and multi-agent debate
  125. Quantizing With Randomized Hadamard Transforms: Efficient Heuristic Now ProvenEfficient LLM inference and serving → KV cache quantization
  126. Adding Thermal Awareness to Visual Systems in Real-Time via Distilled Diffusion ModelsDiffusion models → Image generation, text-to-image and editing
  127. 4DThinker: Thinking with 4D Imagery for Dynamic Spatial UnderstandingLLM reasoning → Spatial and embodied reasoning +2
  128. Safety Anchor: Defending Harmful Fine-tuning via Geometric BottlenecksLLM alignment / RLHF / preference optimisation → Safety alignment and jailbreak attacks +1
  129. Verifiable Model-Free Safety Filters via Reinforcement LearningDeep reinforcement learning methods → Safe and constrained RL
  130. BioResearcher: Scenario-Guided Multi-Agent for Translational MedicineLLM agents and tool use → Medical and scientific domain agents
  131. Towards Steering without Sacrifice: Principled Training of Steering Vectors for Prompt-only InterventionsMechanistic interpretability → Activation steering methods
  132. TACT: Mitigating Overthinking and Overacting in Coding Agents via Activation SteeringLLM reasoning → Reasoning for agents, planning and code +2
  133. BehaviorGuard: Online Backdoor Defense for Deep Reinforcement LearningDeep reinforcement learning methods → Adversarial robustness and RL in finance
  134. Physical Fidelity Reconstruction via Improved Consistency-Distilled Flow Matching for Dynamical SystemsDiffusion models → Flow matching, time series and science +1
  135. PragLocker: Protecting Agent Intellectual Property in Untrusted Deployments via Non-Portable PromptsLLM agents and tool use → Agent threats, guardrails and governance
  136. Training Transformers for KV Cache CompressibilityEfficient LLM inference and serving → KV cache optimization
  137. Beyond Uniform Credit Assignment: Selective Eligibility Traces for RLVRLLM reasoning → Reinforcement learning for reasoning
  138. TheraAgent: Self-Improving Therapeutic Agent for Precise and Comprehensive Treatment PlanningLLM agents and tool use → Medical and scientific domain agents +1
  139. Tatarstan Toponyms: A Bilingual Dataset and Hybrid RAG System for Geospatial Question AnsweringRetrieval-augmented generation → Generation and context integration
  140. Plug-and-Play Label Map Diffusion for Universal Goal-Oriented NavigationDiffusion models → Diffusion and flow policies for robotics
  141. HaM-World: Soft-Hamiltonian World Models with Selective Memory for PlanningWorld models → Latent predictive world models +1
  142. Near-Policy: Accelerating On-Policy Distillation via Asynchronous Generation and Selective PackingKnowledge distillation → On-policy and self-distillation for LLMs
  143. ICU-Bench:Benchmarking Continual Unlearning in Multimodal Large Language ModelsMultimodal LLMs / vision-language models → MLLM jailbreaks and adversarial robustness +1
  144. Think, then Score: Decoupled Reasoning and Scoring for Video Reward ModelingLLM reasoning → Multimodal reasoning +1
  145. Plug-and-play Class-aware Knowledge Injection for Prompt Learning with Visual-Language ModelMultimodal LLMs / vision-language models → Prompt learning and adaptation of CLIP
  146. Null Space Constrained Contrastive Visual Forgetting for MLLM UnlearningLLM safety and security → Machine unlearning
  147. VisMMOE: Exploiting Visual-Expert Affinity for Efficient Visual-Language MoE OffloadingMixture-of-experts → Multimodal MoE models +1
  148. Logic-Regularized Verifier Elicits Reasoning from LLMsLLM reasoning → Reasoning benchmarks and evaluation
  149. Beyond Steering Vector: Flow-based Activation Steering for Inference-Time InterventionMechanistic interpretability → Activation steering methods
  150. MTL-MAD: Multi-Task Learners are Effective Medical Anomaly DetectorsMixture-of-experts → Vision mixture-of-experts
  151. RepFlow: Representation Enhanced Flow Matching for Causal Effect EstimationDiffusion models → Diffusion foundations and sampling
  152. DBMSolver: A Training-free Diffusion Bridge Sampler for High-Quality Image-to-Image TranslationDiffusion models → Image generation, text-to-image and editing
  153. MoE-Hub: Taming Software Complexity for Seamless MoE Overlap with Hardware-Accelerated Communication on Multi-GPU SystemsMixture-of-experts → Efficient MoE training and inference systems
  154. Training-Free Dense Hand Contact Estimation with Multi-Modal Large Language ModelsMultimodal LLMs / vision-language models → Embodied and 3D vision-language models
  155. Agentic, Context-Aware Risk Intelligence in the Internet of ValueLLM agents and tool use → Financial trading and forecasting agents
  156. CITE: Anytime-Valid Statistical Inference in LLM Self-ConsistencyLLM reasoning → Test-time compute and efficient thinking
  157. Retain-Neutral Surrogates for Min-Max UnlearningLLM safety and security → Machine unlearning
  158. SkillScope: Toward Fine-Grained Least-Privilege Enforcement for Agent SkillsLLM agents and tool use → Agent threats, guardrails and governance
  159. SOPE: Stabilizing Off-Policy Evaluation for Online RL with Prior DataDeep reinforcement learning methods → Deep RL algorithms and representations
  160. SANEmerg: An Emergent Communication Framework for Semantic-aware Agentic AI NetworkingLLM agents and tool use → General LLM agent frameworks and benchmarks
  161. Gradient-Momentum Coupling: A Parameter-Space Proxy for Learning ProgressDeep reinforcement learning methods → Deep RL algorithms and representations
  162. VideoRouter: Query-Adaptive Dual Routing for Efficient Long-Video UnderstandingMultimodal LLMs / vision-language models → Video understanding with MLLMs
  163. LoopTrap: Termination Poisoning Attacks on LLM AgentsLLM agents and tool use → Agent threats, guardrails and governance +1
  164. Evaluation Awareness in Language Models Has Limited Effect on BehaviourLLM reasoning → Reasoning benchmarks and evaluation
  165. Unifying Scientific Communication: Fine-Grained Correspondence Across Scientific MediaMultimodal LLMs / vision-language models → Chart understanding and anomaly detection
  166. From Chat to Interview: Agentic Requirements Elicitation with an Experience OntologyLLM agents and tool use → Software engineering coding agents
  167. AGPO: Asymmetric Group Policy Optimization for Verifiable Reasoning and Search Ads Relevance at JDLLM reasoning → Reinforcement learning for reasoning
  168. LeakDojo: Decoding the Leakage Threats of RAG SystemsRetrieval-augmented generation → Privacy and security of RAG
  169. Long-Horizon Q-Learning: Accurate Value Learning via n-Step InequalitiesDeep reinforcement learning methods → RL theory and sample complexity
  170. CXR-ContraBench: Benchmarking Negated-Option Attraction in Medical VLMsMultimodal LLMs / vision-language models → Medical vision-language models
  171. Retrieval from Within: An Intrinsic Capability of Attention-Based ModelsRetrieval-augmented generation → RAG robustness and adaptive retrieval
  172. Selective Rollout: Mid-Trajectory Termination for Multi-Sample Agent RLLLM reasoning → Reinforcement learning for reasoning +1
  173. GazeMind: A Gaze-Guided LLM Agent for Personalized Cognitive Load AssessmentLLM agents and tool use → Multimodal and embodied agents
  174. CircuitFormer: A Circuit Language Model for Analog Topology Design from Natural Language PromptMechanistic interpretability → Circuits and mechanistic analysis
  175. X-OmniClaw Technical Report: A Unified Mobile Agent for Multimodal Understanding and InteractionLLM agents and tool use → GUI and mobile agents
  176. BioTool: A Comprehensive Tool-Calling Dataset for Enhancing Biomedical Capabilities of Large Language ModelsLLM agents and tool use → Medical and scientific domain agents
  177. Transformers Provably Implement In-Context Reinforcement Learning with Policy ImprovementDeep reinforcement learning methods → Deep RL algorithms and representations
  178. RVPO: Risk-Sensitive Alignment via Variance RegularizationLLM alignment / RLHF / preference optimisation → LLM preference optimization methods
  179. Multi-Dimensional Behavioral Evaluation of Agentic Stock Prediction Systems Using Large Language Model Judges with Closed-Loop Reinforcement Learning FeedbackLLM agents and tool use → General LLM agent frameworks and benchmarks
  180. ReFlect: An Effective Harness System for Complex Long-Horizon LLM ReasoningLLM reasoning → Test-time compute and efficient thinking
  181. SDFlow: Similarity-Driven Flow Matching for Time Series GenerationDiffusion models → Flow matching, time series and science
  182. SkillRet: A Large-Scale Benchmark for Skill Retrieval in LLM AgentsLLM agents and tool use → Agent skill learning and evolution
  183. Detecting Time Series Anomalies Like an Expert: A Multi-Agent LLM Framework with Specialized AnalyzersLLM agents and tool use → Software engineering coding agents
  184. B3-Net: Controlled Posterior Bridge Learning for Multi-Task Dense PredictionState-space sequence models (S4, Mamba) → Vision Mamba backbones
  185. More Is Not Always Better: Cross-Component Interference in LLM Agent ScaffoldingLLM agents and tool use → General LLM agent frameworks and benchmarks
  186. TriRelVLA: Triadic Relational Structure for Generalizable Embodied ManipulationVision-language-action models → 3D and spatial-aware VLA
  187. Conceal, Reconstruct, Jailbreak: Exploiting the Reconstruction-Concealment Tradeoff in MLLMsMultimodal LLMs / vision-language models → MLLM jailbreaks and adversarial robustness +1
  188. SafeHarbor: Defining Precise Decision Boundaries via Hierarchical Memory-Augmented Guardrail for LLM Agent SafetyLLM agents and tool use → Agent threats, guardrails and governance +1
  189. Active Learning for Communication Structure Optimization in LLM-Based Multi-Agent SystemsLLM agents and tool use → General LLM agent frameworks and benchmarks
  190. Knowledge-Graph Paths as Intermediate Supervision for Self-Evolving Search AgentsLLM agents and tool use → Web, search, and research agents +1
  191. Inference-Time Budget Control for LLM Search AgentsLLM agents and tool use → Web, search, and research agents
  192. When Quantization Is Free: An int4 KV Cache That Outruns fp16 on Apple SiliconEfficient LLM inference and serving → KV cache quantization
  193. Irminsul: MLA-Native Position-Independent Caching for Agentic LLM ServingLLM agents and tool use → Software engineering coding agents +1
  194. GCCM: Enhancing Generative Graph Prediction via Contrastive Consistency ModelDiffusion models → Flow matching, time series and science
  195. R2H-Diff: Guided Spectral Diffusion Model for RGB-to-Hyperspectral ReconstructionDiffusion models → Inverse problems and image restoration
  196. DataDignity: Training Data Attribution for Large Language ModelsLLM safety and security → Membership inference attacks
  197. PersonaTeaming: Supporting Persona-Driven Red-Teaming for Generative AILLM safety and security → Prompt injection and agent security
  198. Chain of Risk: Safety Failures in Large Reasoning Models and Mitigation via Adaptive Multi-Principle SteeringLLM reasoning → Reasoning for agents, planning and code
  199. Large Vision-Language Models Get Lost in AttentionMultimodal LLMs / vision-language models → Visual token pruning and efficiency
  200. Sparse-to-Complete: From Sparse Image Captures to Complete 3D ScenesDiffusion models → 3D, point cloud and scene generation
  201. XL-SafetyBench: A Country-Grounded Cross-Cultural Benchmark for LLM Safety and Cultural SensitivityLLM safety and security → Safety alignment, refusal and guardrails
  202. Retrieval-Conditioned Topology Selection with Provable Budget Conservation for Multi-Agent Code GenerationLLM agents and tool use → Software engineering coding agents
  203. Negative Before Positive: Asymmetric Valence Processing in Large Language ModelsMechanistic interpretability → Activation steering methods
  204. Text-Graph Synergy: A Bidirectional Verification and Completion Framework for RAGLLM reasoning → Knowledge- and tool-augmented reasoning +1
  205. TokenStack: A Heterogeneous HBM-PIM Architecture and Runtime for Efficient LLM InferenceEfficient LLM inference and serving → Serving systems and request scheduling
  206. Architecture Matters: Comparing RAG Systems under Knowledge Base PoisoningRetrieval-augmented generation → Privacy and security of RAG +1
  207. Accelerating MoE with Dynamic In-Switch Computing on Multi-GPUsMixture-of-experts → Efficient MoE training and inference systems
  208. The Cost of Context: Mitigating Textual Bias in Multimodal Retrieval-Augmented GenerationMultimodal LLMs / vision-language models → Document understanding and multimodal RAG +1
  209. Causal Probing for Internal Visual Representations in Multimodal Large Language ModelsMultimodal LLMs / vision-language models → VLM benchmarks and visual reasoning
  210. Operationalizing Ethics for AI Agents: How Developers Encode Values into Repository Context FilesLLM agents and tool use → Software engineering coding agents
  211. Belief Memory: Agent Memory Under Partial ObservabilityLLM agents and tool use → Agent memory systems
  212. AlphaCrafter: Harnessing Multi-Agent Workflows for Cross-Sectional Quantitative TradingLLM agents and tool use → Financial trading and forecasting agents
  213. AstroAlertBench: Evaluating the Accuracy, Reasoning, and Honesty of Multimodal LLMs in Astronomical ClassificationLLM reasoning → Knowledge- and tool-augmented reasoning +1
  214. Nonsense Helps: Prompt Space Perturbation Broadens Reasoning ExplorationLLM reasoning → Reinforcement learning for reasoning
  215. BitCal-TTS: Bit-Calibrated Test-Time Scaling for Quantized Reasoning ModelsLLM reasoning → Test-time compute and efficient thinking
  216. FedeKD: Energy-Based Gating for Robust Federated Knowledge Distillation under Heterogeneous SettingsKnowledge distillation → Federated, robust and private distillation
  217. A Novel Graph-Regulated Disentangling Mamba Model with Sparse Tokens for Enhanced Tree Species Classification from MODIS Time SeriesState-space sequence models (S4, Mamba) → Hyperspectral image analysis with Mamba
  218. SPARK: Self-Play with Asymmetric Reward from Knowledge GraphsSynthetic data and self-improvement for LLMs → Self-play RL for reasoning
  219. Adaptive Q-Chunking for Offline-to-Online Reinforcement LearningOffline RL and imitation learning from data → Offline RL algorithms
  220. AgenticRAG: Agentic Retrieval for Enterprise Knowledge BasesLLM agents and tool use → Web, search, and research agents
  221. EdgeServing: Deadline-Aware Multi-DNN Serving at the EdgeEfficient LLM inference and serving → Early exit and edge inference
  222. L-FNO: Lorentzian Fourier Neural Operator for Stochastic Event Dynamics
  223. Self-Organising Digital Circuits
  224. MAPS: Modeling Co-Existing Subjective Perspectives and Shared Meaning in Multi-Agent Cognitive Dialogue
  225. Automatically Evolving Prompt Guidelines for Task-Specific Optimization
  226. I'm Sorry, but I Can't Help with Braille: Revealing Accessibility Failures in State-of-the-Art LLMs
  227. SWIFT: Spatio-temporal Wavelet Integrated Forecasting Framework for Workload Traces
  228. Examining Human-Like Behaviors in LLMs: A Multi-Dimensional Analysis of Model Behaviors, User Factors, and System Prompts
  229. The AI Legal Specialist: A Juridically Autonomous Professional Profile for AI Governance
  230. PyCC.id: A package for hypothesis-driven equation discovery with structural identifiability
  231. Grokers: Bottom-Up Inductive Comprehension and Write-Time Intelligence over Typed Knowledge Graphs
  232. Features have life history. And we should care
  233. Orth-Dion: Eliminating Geometric Mismatch in Distributed Low-Rank Spectral Optimization
  234. Vidya: An AI-Driven Modular Pipeline for Archival Automation and Semantic Metadata Enrichment
  235. Decentralized Conformal Novelty Detection via Quantized Model Exchange
  236. SLayerGen: a Crystal Generative Model for all Space and Layer Groups
  237. Can LLMs Predict Polymer Physics Just by Reading Synthesis and Processing Prose?
  238. Dimensional Coactivation for Representational Consistency in Frozen Vision Foundation Models
  239. LLM Translation of Compiler Intermediate Representation
  240. Smart Railway Obstruction Detection System using IoT and Computer Vision
  241. GPU-Accelerated Synthesis of Mixed-Boolean Arithmetic: Beyond Caching
  242. An Explainable Unsupervised-to-Supervised Machine Learning Framework for Dietary Pattern Discovery Using UK National Dietary Survey Data
  243. Resource-Aware Evolutionary Neural Architecture Search for Cardiac MRI Segmentation
  244. Distributional Spectral Diagnostics for Localizing Grokking Transitions
  245. Unlocking High-Fidelity Molecular Generation from Mass Spectra via Dual-Stream Line Graph Diffusion
  246. An Interpretable and Scalable Framework for Evaluating Large Language Models
  247. Dr-BA: Separable Optimization for Direct Radar Bundle Adjustment & Localization
  248. Learning Material-Aware Hamiltonian Risk Fields for Safe Navigation
  249. Beyond the Wrapper: Identifying Artifact Reliance in Static Malware Classifiers using TRUSTEE
  250. BGM-IV: AI-Powered Bayesian Generative Modeling for Instrumental Variable Regression with High-Dimensional Covariates
  251. Learning Cross-Atlas Consistent Brain Disorder Representations via Disentangled Multi-Atlas Functional Connectivity Learning
  252. Delulu: A Verified Multi-Lingual Benchmark for Code Hallucination Detection in Fill-in-the-Middle Tasks
  253. Semantic Prior Guided One-View 6D Pose Estimation for Novel Objects
  254. Self-Driving Datasets: From 20 Million Papers to Nuanced Biomedical Knowledge at Scale
  255. FlashMol: High-Quality Molecule Generation in as Few as Four Steps
  256. Inductive Power Grid Cascading Failure Analysis with GRU-Gated Graph Attention
  257. Equivalence of Coarse and Fine-Grained Models for Learning with Distribution Shift
  258. Adaptive auditing of AI systems with anytime-valid guarantees
  259. Optimal Experiments for Partial Causal Effect Identification
  260. TrajGANR: Trajectory-Centric Urban Multimodal Learning via Geospatially Aligned Neural Representations
  261. Drawing Lines in Psychological Space: What K-means Clustering Reveals in Simulated and Real Psychometric Data
  262. FastOmniTMAE: Parallel Clause Learning for Scalable and Hardware-Efficient Tsetlin Embeddings
  263. Bridging Textual Profiles and Latent User Embeddings for Personalization
  264. Group of Skills: Group-Structured Skill Retrieval for Agent Skill Libraries
  265. A Differentiable Bayesian Relaxation for Latent Partial-Order Inference
  266. Decentralized Time-Varying Optimization for Streaming Data via Temporal Weighting
  267. AI and Consciousness: Shifting Focus Towards Tractable Questions
  268. Locally Near Optimal Piecewise Linear Regression in High Dimensions via Difference of Max-Affine Functions
  269. Kurtosis-Guided Denoising Score Matching for Tabular Anomaly Detection
  270. Rollback-Free Stable Brick Structures Generation
  271. Adaptive Memory Decay for Log-Linear Attention
  272. ProtoSSL: Self-Supervised Pretraining and Downstream Transfer for Projection-Based Prototype Models
  273. Constrained Triple-Head Price Optimization (C3PO) Network for Discrete-Choice Models
  274. MultiSoc-4D: A Benchmark for Diagnosing Instruction-Induced Label Collapse in Closed-Set LLM Annotation of Bengali Social Media
  275. Bias and Uncertainty in LLM-as-a-Judge Estimation
  276. A Nonlinear Singular Value Theory for Neural Networks
  277. A Reproducible Optimisation Protocol for Calibrating Prompt-Based Large Language Model Workflows in Evidence Synthesis
  278. Learned Lyapunov Shielding for Adaptive Control
  279. MAGIQ: A Post-Quantum Multi-Agentic AI Governance System with Provable Security
  280. Target-Aware Data Augmentation for SAT Prediction
  281. XiYOLO: Energy-Aware Object Detection via Iterative Architecture Search and Scaling
  282. In-Context Credit Assignment via the Core
  283. Generalising Travel Time Prediction To Varying Route Choices In Urban Networks
  284. Tyche: One Step Flow for Efficient Probabilistic Weather Forecasting
  285. LLMs are not (consistently) Bayesian: Quantifying internal (in)consistencies of LLMs' probabilistic beliefs
  286. You Only Stack Once (YOSO): A Motion-Filtered, Deep-Learning Framework for Detecting Faint Moving Sources
  287. Advancing Reliable Synthetic Video Detection: Insights from the SAFE Challenge
  288. Dual-Scale Temporal Fusion Reveals Structured Predictability in Subseasonal-to-Seasonal Temperature Prediction
  289. Conservative Flows: A New Paradigm of Generative Models
  290. MELD: Multi-Task Equilibrated Learning Detector for AI-Generated Text
  291. Streaming Adversarial Robustness in Fuzzy ARTMAP: Mechanism-Aligned Evaluation, Progressive Training, and Interpretable Diagnostics
  292. Reflections and New Directions for Human-Centered Large Language Models
  293. Accelerated Relax-and-Round for Concave Coverage Problems
  294. McNdroid: A Longitudinal Multimodal Benchmark for Robust Drift Detection in Android Malware
  295. Towards Fairness under Label Bias in Image Segmentation: Impact, Measurement and Mitigation
  296. TriDE: Triangle-Consistent Translation Directions for Global Camera Pose Estimation
  297. TajPersLexon: A Tajik-Persian Lexical Resource and Hybrid Model for Cross-Script Low-Resource NLP
  298. Muon with Nesterov Momentum: Heavy-Tailed Noise and (Randomized) Inexact Polar Decomposition
  299. Kernel Selection is Model Selection: A Unified Complexity-Penalized Approach for MMD Two-Sample Tests
  300. Better Protein Function Prediction by Modeling Survivorship Bias
  301. Temporal Attention for Adaptive Control of Euler-Lagrange Systems with Unobservable Memory
  302. AdpSplit: Error-Driven Adaptive Splitting for Faster Geometry Discovery in 3D Gaussian Splatting
  303. EULER-ADAS: Energy-Efficient & SIMD-Unified Logarithmic-Posit Engine for Precision-Reconfigurable Approximate ADAS Acceleration
  304. One Operator for Many Densities: Amortized Approximation of Conditioning by Neural Operators
  305. Continuous First, Discrete Later: VQ-VAEs Without Dimensional Collapse
  306. When Descent Is Too Stable: Event-Triggered Hamiltonian Learning to Optimize
  307. A Sharp Finite-Iteration Theory for Asynchronous Categorical Distributional Temporal-Difference Learning
  308. Dataset Watermarking with Provable Black-Box Detection
  309. Multi-Objective Multi-Agent Bandits: From Learning Efficiency to Fairness Optimization
  310. Nonparametric estimation of time-varying network connections by multi-stage smoothing
  311. Knowledge Transfer Scaling Laws for 3D Medical Imaging
  312. Benchmarked Yet Not Measured -- Generative AI Should be Evaluated Against Real-World Utility
  313. LLM-Guided Open Hypothesis Learning from Autonomous Scanning Probe Microscopy Experiments
  314. On Privacy in Data-Space Tabular Diffusion Models: Influential Factors, Attacker Knowledge, and Metrics
  315. Attribution-Based Neuron Utility for Plasticity Restoration in Deep Networks
  316. PAMPOS: Causal Transformer-based Trajectory Prediction for Attack-Agnostic Misbehavior Detection in V2X Networks
  317. IntentGrasp: A Comprehensive Benchmark for Intent Understanding
  318. ProtSent: Protein Sentence Transformers
  319. How Does Attention Help? Insights from Random Matrices on Signal Recovery from Sequence Models
  320. A Rod Flow Model for Adam at the Edge of Stability
  321. Overcoming data scarcity through multi-center federated learning for organs-at-risk segmentation in pediatric upper abdominal radiotherapy
  322. Uneven Evolution of Cognition Across Generations of Generative AI Models
  323. Model-to-Data Distillation for Graph Neural Networks
  324. Enhancing Eye Movement Biometrics for User Authentication via Continuous Gaze Offset Score Fusion
  325. LookWhen? Fast Video Recognition by Learning When, Where, and What to Compute
  326. MIND: Monge Inception Distance for Generative Models Evaluation
  327. VITA-QinYu: Expressive Spoken Language Model for Role-Playing and Singing
  328. A Linear-Transformer Hybrid for SNP-Based Genotype-to-Phenotype Prediction in Grapevine
  329. Physics-based Digital Twins for Integrated Thermal Energy Systems Using Active Learning
  330. A Statistical Framework for Algorithmic Collective Action with Multiple Collectives
  331. HumanNet: Scaling Human-centric Video Learning to One Million Hours
  332. A Closed-Form Upper Bound for Admissible Learning-Rate Steps in Belief-Space Dynamics
  333. Geometric Kolmogorov--Arnold Network (GeoKAN)
  334. Trust Without Trusting: A Recomputable Trust Protocol for Autonomous Agents
  335. STDA-Net: Spectrogram-Based Domain Adaptation for cross-dataset Sleep Stage Classification
  336. Gated QKAN-FWP: Scalable Quantum-inspired Sequence Learning
  337. Beyond Factor Aggregation: Gauge-Aware Low-Rank Server Representations for Federated LoRA
  338. Semantic State Abstraction Interfaces for LLM-Augmented Portfolio Decisions: Multi-Axis News Decomposition and RL Diagnostics
  339. The EDelta-MHC-Geo Transformer: Adaptive Geodesic Operations with Guaranteed Orthogonality
  340. Medical Imaging Classification with Cold-Atom Reservoir Computing using Auto-Encoders and Surrogate-Driven Training
  341. Transformer-Based Wildlife Species Classification from Daily Movement Trajectories
  342. Enabling Unsupervised Training of Deep EEG Denoisers With Intelligent Partitioning
  343. TUANDROMD-X: Advanced Entropy and Visual Analytics Dataset for Enhanced Malware Detection and Classification
  344. Edge Deep Learning in Computer Vision and Medical Diagnostics: A Comprehensive Survey
  345. ActCam: Zero-Shot Joint Camera and 3D Motion Control for Video Generation
  346. BAMI: Training-Free Bias Mitigation in GUI Grounding
  347. Why Global LLM Leaderboards Are Misleading: Small Portfolios for Heterogeneous Supervised ML
  348. Optimizer-Model Consistency: Full Finetuning with the Same Optimizer as Pretraining Forgets Less
  349. Inductive Venn-Abers and related regressors
  350. Edge-specific signal propagation on mature chromophore-region 3D mechanism graphs for fluorescent protein quantum-yield prediction
  351. Are We Making Progress in Multimodal Domain Generalization? A Comprehensive Benchmark Study
  352. GlazyBench: A Benchmark for Ceramic Glaze Property Prediction and Image Generation
  353. Concept-Based Abductive and Contrastive Explanations for Behaviors of Vision Models
  354. DPM++: Dynamic Masked Metric Learning for Occluded Person Re-identification
  355. Crafting Reversible SFT Behaviors in Large Language Models
  356. Hybrid Quantum-Classical GANs for the Generation of Adversarial Network Flows
  357. LiVeAction: a Lightweight, Versatile, and Asymmetric Neural Codec Design for Real-time Operation
  358. PianoCoRe: Combined and Refined Piano MIDI Dataset
  359. Parser agreement and disagreement in L2 Korean UD: Implications for human-in-the-loop annotation
  360. Algospeak, Hiding in the Open: The Trade-off Between Legible Meaning and Detection Avoidance
  361. When and Why SignSGD Outperforms SGD: A Theoretical Study Based on ℓ1-norm Lower Bounds
  362. Online Bayesian Calibration under Gradual and Abrupt System Changes
  363. The Structural Origin of Attention Sink: Variance Discrepancy, Super Neurons, and Dimension Disparity
  364. Transformers Efficiently Perform In-Context Logistic Regression via Normalized Gradient Descent
  365. DARTS: Targeting Prognostic Covariates in Budget-Constrained Sequential Experiments
  366. A Geometry-Aware Residual Correction of Hagan's SABR Implied Volatility Formula
  367. Weight-Decay Turns Transformer Loss Landscapes Villani: Functional-Analytic Foundations for Optimization and Generalization
  368. Cross-Modal Navigation with Multi-Agent Reinforcement Learning
  369. Automated Clinical Report Generation for Remote Cognitive Remediation: Comparing Knowledge-Engineered Templates and LLMs in Low-Resource Settings
  370. BRICKS: Compositional Neural Markov Kernels for Zero-Shot Radiation-Matter Simulation
  371. Towards Metric-Faithful Neural Graph Matching
  372. Learning Distributionally Robust First-Order Methods for Convex Optimization
  373. PairAlign: A Framework for Autoregressive Tokenization via Self-Alignment with Applications to Audio Tokenization
  374. On the Safety of Graph Representation Learning
  375. Directional Consistency as a Complementary Optimization Signal: The GONO Framework
  376. Solving Minimal Problems Without Matrix Inversion Using FFT-Based Interpolation
  377. CLAD: A Clustered Label-Agnostic Federated Learning Framework for Joint Anomaly Detection and Attack Classification
  378. SNAPO: Smooth Neural Adjoint Policy Optimization for Optimal Control via Differentiable Simulation
  379. Dynamic Treatment on Networks
  380. Criticality and Saturation in Orthogonal Neural Networks
  381. Feature Dimensionality Outweighs Model Complexity in Breast Cancer Subtype Classification Using TCGA-BRCA Gene Expression Data
  382. Optimal Counterfactual Search in Tree Ensembles: A Study Across Modeling and Solution Paradigms
  383. Coordination Matters: Evaluation of Cooperative Multi-Agent Reinforcement Learning
  384. Long Context Pre-Training with Lighthouse Attention
  385. Efficient Pre-Training with Token Superposition
  386. Hedging Memory Horizons for Non-Stationary Prediction via Online Aggregation
  387. Ex Ante Evaluation of AI-Induced Idea Diversity Collapse
  388. Sparkle: Realizing Lively Instruction-Guided Video Background Replacement via Decoupled Guidance
  389. SpatialEpiBench: Benchmarking Spatial Information and Epidemic Priors in Forecasting
  390. Market-Alignment Risk in Pricing Agents: Trace Diagnostics and Trace-Prior RL under Hidden Competitor State
  391. Process Matters more than Output for Distinguishing Humans from Machines
  392. Optimizing Social Utility in Sequential Experiments
  393. Efficient Techniques for Data Reconstruction, with Finite-Width Recovery Guarantees
  394. Is One Layer Enough? Understanding Inference Dynamics in Tabular Foundation Models
  395. Beyond Reproducibility: Towards Security-Aware Evaluation of Research Artifacts
  396. The Frequency Confound in Language-Model Surprisal and Metaphor Novelty
  397. PACZero: PAC-Private Fine-Tuning of Language Models via Sign Quantization
  398. Cubit: Token Mixer with Kernel Ridge Regression
  399. 3D MRI Image Pretraining via Controllable 2D Slice Navigation Task
  400. Litespark Inference For CPUs: Ultra-Fast SIMD Framework for Ternary (1.58-bit) Language Models
  401. Estimate Level Adjustment For Inference With Proxies Under Random Distribution Shifts
  402. Risk-Controlled Post-Processing of Decision Policies
  403. Towards Emotion Consistency Analysis of Large Language Models in Emotional Conversational Contexts
  404. Probabilistic Dating of Historical Manuscripts via Evidential Deep Regression on Visual Script Features
  405. Q-MMR: Off-Policy Evaluation via Recursive Reweighting and Moment Matching
  406. Dynamic Controlled Variables Based Dynamic Self-Optimizing Control
  407. No Triangulation Without Representation: Generalization in Topological Deep Learning
  408. Diversity Curves for Graph Representation Learning
  409. Measuring Structure in Graph Benchmark Datasets Using Graph Invariants
  410. MINER: Mining Multimodal Internal Representation for Efficient Retrieval
  411. Invariant Features in Language Models: Geometric Characterization and Model Attribution
  412. ORTHOBO: Orthogonal Bayesian Hyperparameter Optimization
  413. FedFrozen: Two-Stage Federated Optimization via Attention Kernel Freezing
  414. Hyperbolic Concept Bottleneck Models
  415. From Review to Design: Ethical Multimodal Driver Monitoring Systems for Risk Mitigation, Incident Response, and Accountability in Automated Vehicles
  416. Neural-Actuarial Longevity Forecasting: Anchoring LSTMs for Explainable Risk Management
  417. COVID-19 Infodemic. Understanding content features in detecting fake news using a machine learning approach
  418. Federated Layer-Wise Exchange for Cross-Client Money Laundering Pattern Detection
  419. From 124 Million Tokens to 1,021 Neologisms: A Large-Scale Pipeline for Automatic Neologism Detection
  420. MiA-Signature: Approximating Global Activation for Long-Context Understanding
  421. Decoupled PFNs: Identifiable Epistemic-Aleatoric Decomposition via Structured Synthetic Priors
  422. WavCube: Unifying Speech Representation for Understanding and Generation via Semantic-Acoustic Joint Modeling
  423. FRInGe: Distribution-Space Integrated Gradients with Fisher--Rao Geometry
  424. GATHER: Convergence-Centric Hyper-Entity Retrieval for Zero-Shot Cell-Type Annotation
  425. SparseForge: Efficient Semi-Structured LLM Sparsification via Annealing of Hessian-Guided Soft-Mask
  426. Consistent Geometric Deep Learning via Hilbert Bundles and Cellular Sheaves
  427. Lecture Notes on Statistical Physics and Neural Networks
  428. Automated alignment is harder than you think
  429. Covariate Balancing and Riesz Regression Should Be Guided by the Neyman Orthogonal Score in Debiased Machine Learning
  430. Data-Driven Covariate Selection for Nonparametric and Cycle-Agnostic Causal Effect Estimation
  431. MinMax Recurrent Neural Cascades
  432. Rethinking Vacuity for OOD Detection in Evidential Deep Learning
  433. Empirical Evidence for Simply Connected Decision Regions in Image Classifiers
  434. Debiased Multimodal Personality Understanding through Dual Causal Intervention
  435. eXplaining to Learn (eX2L): Regularization Using Contrastive Visual Explanation Pairs for Distribution Shifts
  436. Preliminary Insights in Chronos Frequency Data Understanding and Reconstruction
  437. The frame-level leakage trap: rethinking evaluation protocols for intrinsic image decomposition, with source-separable uncertainty as a case study
  438. Memory Efficient Full-gradient Attacks (MEFA) Framework for Adversarial Defense Evaluations
  439. SwiftI2V: Efficient High-Resolution Image-to-Video Generation via Conditional Segment-wise Generation
  440. Order-Agnostic Autoregressive Modelling with Missing Data
  441. SEQUOR: A Multi-Turn Benchmark for Realistic Constraint Following
  442. Topological Signatures of Grokking
  443. Is Escalation Worth It? On the Depth of LLM Cascades
  444. Human-AI Co-Evolution and Epistemic Collapse: A Dynamical Systems Perspective
  445. Prediction and Empowerment: A Theory of Agency through Bridge Interfaces
  446. Mind the Gap? A Distributional Comparison of Real and Synthetic Priors for Tabular Foundation Models
  447. CoupleEvo: Evolving Heuristics for Coupled Optimization Problems Using Large Language Models
  448. A Benchmark for Strategic Auditee Gaming Under Continuous Compliance Monitoring
  449. Eliciting associations between clinical variables from LLMs via comparison questions across populations
  450. TinyBayes: Closed-Form Bayesian Inference via Jacobi Prior for Real-Time Image Classification on Edge Devices
  451. LINC: Decoupling Local Consequence Scoring from Hidden Matching in Constructive Neural Routing
  452. Fine-Tuning Small Language Models for Solution-Oriented Windows Event Log Analysis
  453. Measuring Evaluation-Context Divergence in Open-Weight LLMs: A Paired-Prompt Protocol with Pilot Evidence of Alignment-Pipeline-Specific Heterogeneity
  454. Gaming the Metric, Not the Harm: Certifying Safety Audits against Strategic Platform Manipulation
  455. SMolLM: Small Language Models Learn Small Molecular Grammar
  456. Who and What? Using Linguistic Features and Annotator Characteristics to Analyze Annotation Variation
  457. NavOne: One-Step Global Planning for Vision-Language Navigation on Top-Down Maps
  458. Pro-KLShampoo: Projected KL-Shampoo with Whitening Recovered by Orthogonalization
  459. End-to-End Identifiable and Consistent Recurrent Switching Dynamical Systems
  460. When Does ℓ2-Boosting Overfit Benignly? High-Dimensional Risk Asymptotics and the ℓ1 Implicit Bias
  461. Perceive, Route and Modulate: Dynamic Pattern Recalibration for Time Series Forecasting
  462. MultiLinguahah : A New Unsupervised Multilingual Acoustic Laughter Segmentation Method
  463. Addressing Labelled Data Scarcity: Taxonomy-Agnostic Annotation of PII Values in HTTP Traffic using LLMs
  464. Region Seeding via Pre-Activation Regularization: A Geometric View of Piecewise Affine Neural Networks
  465. The Metagame of Interpretability and Meta-Attributions
  466. Log-Likelihood, Simpson's Paradox, and the Detection of Machine-Generated Text
  467. Data Language Models: A New Foundation Model Class for Tabular Data
  468. Multimodal Deep Generative Model for Semi-Supervised Learning under Class Imbalance
  469. A Topological Sorting Criterion for Random Causal Directed Acyclic Graphs
  470. Quantifying the Statistical Effect of Rubric Modifications on Human-Autorater Agreement
  471. INEUS: Iterative Neural Solver for High-Dimensional PIDEs
  472. Correct Code, Vulnerable Dependencies: A Large Scale Measurement Study of LLM-Specified Library Versions
  473. PACE: Prune-And-Compress Ensemble Models
  474. Linear Semantic Segmentation for Low-Resource Spoken Dialects
  475. When Labels Have Structure: Improving Image Classification with Hierarchy-Aware Cross-Entropy
  476. On-Orbit Real-Time Wildfire Detection Under On-Board Constraints
  477. ZScribbleSeg: A comprehensive segmentation framework with modeling of efficient annotation and maximization of scribble supervision
  478. ConquerNet: Convolution-Smoothed Quantile ReLU Neural Networks with Minimax Guarantees
  479. Inference-Time Refinement Closes the Synthetic-Real Gap in Tabular Diffusion
  480. Beyond Rigid Alignment: Graph Federated Learning via Dual Manifold Calibration
  481. Trade-off Functions for DP-SGD with Subsampling based on Random Allocation: Tight Upper and Lower Bounds
  482. The Weight Gram Matrix Captures Sequential Feature Linearization in Deep Networks
  483. The Role of Node Features in Graph Pooling
  484. Structure-Preserving Gaussian Processes Via Discrete Euler-Lagrange Equations
  485. Cumulative-Goodness Free-Riding in Forward-Forward Networks: Real, Repairable, but Not Accuracy-Dominant
  486. When Graph Language Models Go Beyond Memorization
  487. Band Together: Untargeted Adversarial Training with Multimodal Coordination against Evasion-based Promotion Attacks
  488. OBLIQ-Bench: Exposing Overlooked Bottlenecks in Modern Retrievers with Latent and Implicit Queries
  489. YEZE at SemEval-2026 Task 9: Detecting Multilingual, Multicultural and Multievent Online Polarization via Heterogeneous Ensembling
  490. Look Beyond Saliency: Low-Attention Guided Dual Encoding for Video Semantic Search
  491. Price of Fairness in Short-Term and Long-Term Algorithmic Selections
  492. A Versatile AI Agent for Rare Disease Diagnosis and Risk Gene Prioritization
  493. ProCompNav: Proactive Instance Navigation with Comparative Judgment for Ambiguous User Queries
  494. AffineLens: Capturing the Continuous Piecewise Affine Functions of Neural Networks
  495. TIDE: Every Layer Knows the Token Beneath the Context
  496. Pushing the accuracy of on-top functionals with agent-driven supervised learning
  497. Differentiable Adaptive 4D Structured Illumination for Joint Capture of Shape and Reflectance
  498. Contrastive Identification and Generation in the Limit
  499. Super-Level-Set Regression: Conditional Quantiles via Volume Minimization
  500. Taming the Entropy Cliff: Variable Codebook Size Quantization for Autoregressive Visual Generation
  501. When Does Trimming Help Conformal Prediction? A Retained-Law Diagnostic under Calibration Contamination
  502. Bandit Learning in General Open Multi-agent Systems
  503. Towards Annotation-Free Validation of MLLMs: A Vision-Language Logical Consistency Metric
  504. Bridging visual saliency and large language models for explainable deep learning in medical imaging
  505. The Granularity Axis: A Micro-to-Macro Latent Direction for Social Roles in Language Models
  506. Systematic Evaluation of Large Language Models for Post-Discharge Clinical Action Extraction
  507. Constrained Contextual Bandits with Adversarial Contexts
  508. Predictive-Generative Drift Decomposition for Speech Enhancement and Separation
  509. In-Context Black-Box Optimization with Unreliable Feedback
  510. Teaching LLMs Program Semantics via Symbolic Execution Traces
  511. Rethinking Adapter Placement: A Dominant Adaptation Module Perspective
  512. SuperFace: Preference-Aligned Facial Expression Estimation Beyond Pseudo Supervision
  513. Expressivity of Bi-Lipschitz Normalizing Flows: A Score-Based Diffusion Perspective
  514. DynT2I-Eval: A Dynamic Evaluation Framework for Text-to-Image Models
  515. One Algorithm, Two Goals: Dual Scoring for Parameter and Data Selection in LLM Fine-Tuning
  516. Beyond Forgetting in Continual Medical Image Segmentation: A Comprehensive Benchmark Study
  517. Graphlets as Building Blocks for Structural Vocabulary in Knowledge Graph Foundation Models
  518. Grokking or Glitching? How Low-Precision Drives Slingshot Loss Spikes
  519. Learning Discrete Autoregressive Priors with Wasserstein Gradient Flow
  520. AI-Generated Images: What Humans and Machines See When They Look at the Same Image
  521. IRC-Bench: Recognizing Entities from Contextual Cues in First-Person Reminiscences
  522. Matrix-Valued Optimism is Matrix-Valued Augmentation: Additive Hybrid Designs for Constrained Optimization
  523. SymDrift: One-Shot Generative Modeling under Symmetries
  524. Autoregressive Visual Generation Needs a Prologue
  525. BUILD-AND-FIND: An Effort-Aware Protocol for Evaluating Agent-Managed Codebases
  526. Diffusion model for SU(N) gauge theories
  527. Continuous Expert Assembly: Instance-Conditioned Low-Rank Residuals for All-in-One Image Restoration
  528. Back to the Beginning of Heuristic Design: Bridging Code and Knowledge with LLMs
  529. BoostLLM: Boosting-inspired LLM Fine-tuning for Few-shot Tabular Classification
  530. Schedule-and-Calibrate: Utility-Guided Multi-Task Reinforcement Learning for Code LLMs
  531. Shallow Prefill, Deep Decoding: Efficient Long-Context Inference via Layer-Asymmetric KV Visibility
  532. CredibleDFGO: Differentiable Factor Graph Optimization with Credibility Supervision
  533. Metonymy in vision models undermines attention-based interpretability
  534. Boosting Self-Supervised Tracking with Contextual Prompts and Noise Learning
  535. Time-Inhomogeneous Preconditioned Langevin Dynamics
  536. OpenGaFF: Open-Vocabulary Gaussian Feature Field with Codebook Attention
  537. Safety Certification is Classification
  538. LARGO: Low-Rank Hypernetwork for Handling Missing Modalities
  539. AMIEOD: Adaptive Multi-Experts Image Enhancement for Object Detection in Low-Illumination Scenes
  540. Revisiting Uncertainty: On Evidential Learning for Partially Relevant Video Retrieval
  541. PoTAcc: A Pipeline for End-to-End Acceleration of Power-of-Two Quantized DNNs
  542. Fast Gauss-Newton for Multiclass Cross-Entropy
  543. MSD-Score: Multi-Scale Distributional Scoring for Reference-Free Image Caption Evaluation
  544. PRISM: Iterative Cross-Modal Posterior Refinement for Dynamic Text-Attributed Graphs
  545. Normalized Architectures are Natively 4-Bit
  546. PersonaGesture: Single-Reference Co-Speech Gesture Personalization for Unseen Speakers
  547. Geometry-Aware Simplicial Message Passing
  548. Correcting heterogeneous diagnostic bias when developing clinical prediction models using causal hidden Markov models
  549. Visual Fingerprints for LLM Generation Comparison
  550. Towards Generation-Efficient Uncertainty Estimation in Large Language Models
  551. RealCam: Real-Time Novel-View Video Generation with Interactive Camera Control
  552. Learning beyond Site Bias for OOD Generalization in Brain Networks
  553. TFM-Retouche: A Lightweight Input-Space Adapter for Tabular Foundation Models
  554. CRISP: Compositional Relations as Invariant Structural Priors for Domain Generalization
  555. Quantum Kernels for Audio Deepfake Detection Using Spectrogram Patch Features
  556. When AI Meets Science: Research Diversity, Interdisciplinarity, Visibility, and Retractions across Disciplines in a Global Surge
  557. More Aligned, Less Diverse? Analyzing the Grammar and Lexicon of Two Generations of LLMs
  558. Pathways to AGI
  559. Strat-LLM: Stratified Strategy Alignment for LLM-based Stock Trading with Real-time Multi-Source Signals
  560. PlotPick: AI-powered batch extraction of numerical data from scientific figures
  561. Matrix-Decoupled Concentration for Autoregressive Sequences: Dimension-Free Guarantees for Sparse Long-Context Rewards
  562. T2I-VeRW: Part-level Fine-grained Perception for Text-to-Image Vehicle Retrieval
  563. PersonaKit (PK): A Plug-and-Play Platform for User Testing Diverse Roles in Full-Duplex Dialogue
  564. From Articles to Premises: Building PrimeFacts, an Extraction Methodology and Resource for Fact-Checking Evidence
  565. Neuromorphic visual attention for Sign-language recognition on SpiNNaker
  566. A Fine-Grained Understanding of Uniform Convergence for Halfspaces
  567. Gaussian mixture models in Hilbert spaces via kernel methods
  568. DiBA: Diagonal and Binary Matrix Approximation for Neural Network Weight Compression
  569. TabCF: Distributional Control Function Estimation with Tabular Foundation Models
  570. iPhoneBlur: A Difficulty-Stratified Benchmark for Consumer Device Motion Deblurring
  571. Prompt-Free and Efficient SAM2 Adaptation for Biomedical Semantic Segmentation via Dual Adapters
  572. Towards Reliable LLM Evaluation: Correcting the Winner's Curse in Adaptive Benchmarking
  573. Sharper Guarantees for Misspecified Kernelized Bandit Optimization
  574. Uncertainty Estimation via Hyperspherical Confidence Mapping
  575. From Coordinate Matching to Structural Alignment: Rethinking Prototype Alignment in Heterogeneous Federated Learning
  576. Temporal Smoothness Doubly Robust Learning for Debiased Knowledge Tracing
  577. Knowing but Not Correcting: Routine Task Requests Suppress Factual Correction in LLMs
  578. TableVista: Benchmarking Multimodal Table Reasoning under Visual and Structural Complexity
  579. Hallucination as an Anomaly: Dynamic Intervention via Probabilistic Circuits
  580. Lightweight Stylistic Consistency Profiling: Robust Detection of LLM-Generated Textual Content for Multimedia Moderation
  581. MAS-Algorithm: A Workflow for Solving Algorithmic Programming Problems with a Multi-Agent System
  582. MobileEgo Anywhere: Open Infrastructure for long horizon egocentric data on commodity hardware
  583. Architecture Shape Governs QNN Trainability: Jacobian Null Space Growth and Parameter Efficiency
  584. RAWild: Sensor-Agnostic RAW Object Detection via Physics-Guided Curve and Grid Modeling
  585. Whole-body CT attenuation and volume charts from routine clinical scans via evidence-grounded LLM report filtering
  586. In Data or Invisible: Toward a Better Digital Representation of Low-Resource Languages with Knowledge Graphs
  587. Which Are the Low-Resource Languages of the Semantic Web?
  588. Backdoor Mitigation in Object Detection via Adversarial Fine-Tuning
  589. Minimizing Modality Gap from the Input Side: Your Speech LLM Can Be a Prosody-Aware Text LLM
  590. Intentmaking and Sensemaking: Human Interaction with AI-Guided Mathematical Discovery
  591. LLM-Driven Design Space Exploration of FPGA-based Accelerators
  592. Quantum-enhanced Large Language Models on Quantum Hardware via Cayley Unitary Adapters
  593. Wisteria: A Unified Multi-Scale Feature Learning Framework for DNA Language Model
  594. From Drops to Grid: Noise-Aware Spatio-Temporal Neural Process for Rainfall Estimation
  595. PREFER: Personalized Review Summarization with Online Preference Learning
  596. Architecture-agnostic Lipschitz-constant Bayesian header and its application to resolve semantically proximal classification errors with vision transformers
  597. Quadratic Objective Perturbation: Curvature-Based Differential Privacy
  598. Understanding Cross-Language Transfer Improvements in Low-Resource HTR: The Role of Sequence Modeling
  599. VARS-FL: Validation-Aligned Client Selection for Non-IID Federated Learning in IoT Systems
  600. Detecting AI-Generated Videos with Spiking Neural Networks
  601. Tuning Derivatives for Causal Fairness in Machine Learning
  602. 3DSS: 3D Surface Splatting for Inverse Rendering
  603. QuadraSHAP: ε-Exact Shapley Values for Product Games in Logarithmic Parallel Time
  604. XDecomposer: Learning Prior-Free Set Decomposition for Multiphase X-ray Diffraction
  605. InkDiffuser: High-Fidelity One-shot Chinese Calligraphy via Differentiable Morphological Optimization
  606. Do Neural Operators Forget Geometry? The Forgetting Hypothesis in Deep Operator Learning
  607. Offline Reinforcement Learning for Rotation Profile Control in Tokamaks
  608. Bridging Passive and Active: Enhancing Conversation Starter Recommendation via Active Expression Modeling
  609. AirQualityBench: A Realistic Evaluation Benchmark for Global Air Quality Forecasting
  610. Hypothesis generation and updating in large language models
  611. Align3D-AD: Cross-Modal Feature Alignment and Dual-Prompt Learning for Zero-shot 3D Anomaly Detection
  612. Taklif.AI: LLM-Powered Platform for Interest-Based Personalized College Assignments
  613. MDN: Parallelizing Stepwise Momentum for Delta Linear Attention
  614. On the Role of Language Representations in Auto-Bidding: Findings and Implications
  615. MolRecBench-Wild: A Real-World Benchmark for Optical Chemical Structure Recognition
  616. ChartZero: Synthetic Priors Enable Zero Shot Chart Data Extraction
  617. HCInfer: An Efficient Inference System via Error Compensation for Resource-Constrained Devices
  618. A Testable Certificate for Constant Collapse in Teacher-Guided VAEs
  619. Sheet As Token: A Graph-Enhanced Representation for Multi-Sheet Spreadsheet Understanding
  620. Detecting Changes in Causal Dependence with Kernels and Copulas
  621. Ratio-based Loss Functions
  622. LCC-LLM: Leveraging Code-Centric Large Language Models for Malware Attribution
  623. Na-IRSTD: Enhancing Infrared Small Target Detection via Native-Resolution Feature Selection and Fusion
  624. Reward Shaping and Action Masking for Compositional Tasks using Behavior Trees and LLMs
  625. Revealing Modular Gradient Noise Imbalance in LLMs: Calibrating Adam via Signal-to-Noise Ratio
  626. Fitted Q-Iteration with Continuous Actions and Adaptive Data: Smoothed Coverage and Minimax Rates
  627. Stego Battlefield: Evaluating Image Steganography Attacks and Steganalysis Defenses
  628. Steering Visual Generation in Unified Multimodal Models with Understanding Supervision
  629. Von Neumann Networks
  630. Estimating the Black-box LLM Uncertainty with Distribution-Aligned Adversarial Distillation
  631. HEDP: A Hybrid Energy-Distance Prompt-based Framework for Domain Incremental Learning
  632. The autoPET3 Challenge: Automated Lesion Segmentation in Whole-Body PET/CT unicodex2013 Multitracer Multicenter Generalization
  633. Confidence is the key: how conformal prediction enhances the generative design of permeable peptides
  634. Adaptive Selection of LoRA Components in Privacy-Preserving Federated Learning
  635. Optimal Confidence Band for Kernel Gradient Flow Estimator
  636. Priming, Path-dependence, and Plasticity: Understanding the molding of user-LLM interaction and its implications from (many) chat logs in the wild
  637. iTRIALSPACE: Programmable Virtual Lesion Trials for Controlled Evaluation of Lung CT Models
  638. Full-Spectrum Graph Neural Networks: Expressive and Scalable
  639. MaMi-HOI: Harmonizing Global Kinematics and Local Geometry for Human-Object Interaction Generation
  640. Jointly Learning Structured Representations and Stabilized Affinity for Human Motion Segmentation
  641. Ray-Aware Pointer Memory with Adaptive Updates for Streaming 3D Reconstruction
  642. Evaluating Explainability in Safety-Critical ATR Systems: Limitations of Post-Hoc Methods and Paths Toward Robust XAI
  643. Polarizable atomic multipoles for learning long-range electrostatics
  644. Best Arm Identification in Generalized Linear Bandits via Hybrid Feedback
  645. Fourier Feature Methods for Nonlinear Causal Discovery: FFML Scoring, TRFF Scoring, and FFCI Testing in Mixed Data
  646. Weak-to-Strong Generalization is Nearly Inevitable (in Linear Models)
  647. HyperLens: Quantifying Cognitive Effort in LLMs with Fine-grained Confidence Trajectory
  648. CoMemNet: A Continual Memory Network with Drift-Aware Sampling for Traffic Prediction
  649. CRAFT: Forgetting-Aware Intervention-Based Adaptation for Continual Learning
  650. Knee Osteoarthritis Severity Grading Using Optimized Deep Learning and LLM-Driven Intelligent AI on Computationally Limited Systems
  651. WARP: A Benchmark for Primal-Dual Warm-Starting of Interior-Point Solvers
  652. Auto Research with Specialist Agents Develops Effective and Non-Trivial Training Recipes
  653. Enabling Federated Inference via Unsupervised Consensus Embedding
  654. Decodable but Not Corrected by Fixed Residual-Stream Linear Steering: Evidence from Medical LLM Failure Regimes
  655. EgoEMG: A Multimodal Egocentric Dataset with Bilateral EMG and Vision for Hand Pose Estimation
  656. Closing the Loop: Unified 3D Scene Generation and Immersive Interaction via LLM-RL Coupling
  657. On the Blessing of Pre-training in Weak-to-Strong Generalization
  658. Resolving the bias-precision paradox with stochastic causal representation learning for personalized medicine
  659. Convex-Geometric Error Bounds for Positive-Weight Kernel Quadrature
  660. An Empirical Study of Proactive Coding Assistants in Real-World Software Development
  661. Budgeted Attention Allocation: Cost-Conditioned Compute Control for Efficient Transformers
  662. Adaptive Physical-Facial Representation Fusion via Subject-Invariant Cross-Modal Prompt Tuning for Video-Based Emotion Recognition
  663. Saliency-Aware Regularized Quantization Calibration for Large Language Models
  664. CFE-PPAR: Compression-friendly encryption for privacy-preserving action recognition leveraging video transformers
  665. Attractor Geometry of Transformer Memory: From Conflict Arbitration to Confident Hallucination
  666. Temporal Functional Circuits: From Spline Plots to Faithful Explanations in KAN Forecasting
  667. Spectral Lens: Activation and Gradient Spectra as Diagnostics of LLM Optimization
  668. MotionGRPO: Overcoming Low Intra-Group Diversity in GRPO-Based Egocentric Motion Recovery
  669. Decomposing the Basic Abilities of Large Language Models: Mitigating Cross-Task Interference in Multi-Task Instruct-Tuning
  670. EGA: Adapting Frozen Encoders for Vector Search with Bounded Out-of-Distribution Degradation
  671. From Dual Tracking to Clipping: Provably Faster Distributionally Robust Multi-Objective Optimization
  672. Structural Correspondence and Universal Approximation in Diagonal plus Low-Rank Neural Networks
  673. Information-Preserving Domain Transfer with Unlabeled Data in Misspecified Simulation-Based Inference
  674. The Missing Evaluation Axis: What 10,000 Student Submissions Reveal About AI Tutor Effectiveness
  675. MUSE: Resolving Manifold Misalignment in Visual Tokenization via Topological Orthogonality
  676. AffectSeek: Agentic Affective Understanding in Long Videos under Vague User Queries
  677. Scaling Pretrained Representations Enables Label-Free Out-of-Distribution Detection Without Fine-Tuning
  678. Learning a Delighting Prior for Facial Appearance Capture in the Wild
  679. One Turn Too Late: Learning When to Intervene Against Multi-Turn Malicious Intent
  680. Spherical Flows for Sampling Categorical Data
  681. Leveraging Image Generators to Address Data Scarcity: The Gen4Regen Dataset for Forest Regeneration Mapping
  682. When2Speak: A Dataset for Temporal Participation and Turn-Taking in Multi-Party Conversations for Large Language Models
  683. Quantum Kernels for Parity-Structured Classification: A Hybrid Pipeline
  684. Retrieval of Coastal Biogeochemical Parameters From Near-Surface Hyperspectral Remote Sensing Reflectance Using Physics-Aware Meta-Learning
  685. RAM-H1200: A Unified Evaluation and Dataset on Hand Radiographs for Rheumatoid Arthritis
  686. LLMSpace: Carbon Footprint Modeling for Large Language Model Inference on LEO Satellites
  687. X-Voice: Enabling Everyone to Speak 30 Languages via Zero-Shot Cross-Lingual Voice Cloning
  688. Optimal Contextual Pricing under Agnostic Non-Lipschitz Demand
  689. Variational Smoothing and Inference for SDEs from Sparse Data with Dynamic Neural Flows
  690. Nearly Optimal Attention Coresets
  691. Prober.ai: Gated Inquiry-Based Feedback via LLM-Constrained Personas for Argumentative Writing Development
  692. When Can Voting Help, Hurt, or Change Course? Exact Structure of Binary Test-Time Aggregation
  693. In-Context Positive-Unlabeled Learning
  694. Uncertainty-Guided Edge Learning for Deep Image Regression in Remote Sensing
  695. AeroJEPA: Learning Semantic Latent Representations for Scalable 3D Aerodynamic Field Modeling
  696. A Scalable Digital Twin Framework for Energy Optimization in Data Centers
  697. Accelerating LMO-Based Optimization via Implicit Gradient Transport
  698. Text-to-CAD Retrieval: a Strong Baseline
  699. Stability of the Monge Map in Semi-Dual Optimal Transport
  700. Relaxed Sparsest-Permutation Formulation for Causal Discovery at Scale
  701. Locality-aware Private Class Identification for Domain Adaptation with Extreme Label Shift
  702. Who Prices Cognitive Labor in the Age of Agents? Compute-Anchored Wages
  703. Extremely coarse learning objectives induce human-aligned representations in AI vision models
  704. Characterizing Brazilian Atlantic Forest Restoration Outcomes with Geospatial AlphaEarth Embeddings
  705. Autoregressive One-Step Generative Modeling for Dynamical System Forecasting
  706. Housing Potential Common Data Model and City Digital Twin
  707. Adversarial Graph Neural Network Benchmarks: Towards Practical and Fair Evaluation
  708. A Few Good Clauses: Comparing LLMs vs Domain-Trained Small Language Models on Structured Contract Extraction
  709. Energy Generative Modeling: A Lyapunov-based Energy Matching Perspective
  710. Discrete Elastic Ribbons: A Unified Discrete Differential Geometry Framework for One-Dimensional Energy Models
  711. Anatomy of a Query: W5H Dimensions and FAR Patterns for Text-to-SQL Evaluation