mlsanityData to 14 Oct 2026

Papers

New on arXiv

Sat 25 Apr 2026

178 papers

  1. Dynamic Governance of Multi-LLM Agent Systems for Collaborative Conversational OutcomesLLM agents and tool use → General LLM agent frameworks and benchmarks
  2. GAZE: Grounded Agentic Zero-shot Evaluation with Viewer-Level Tools and Literature Retrieval on Rare Brain MRIMultimodal LLMs / vision-language models → Medical vision-language models
  3. Ulterior Motives: Detecting Misaligned Reasoning in Continuous Thought ModelsLLM reasoning → Prompting and chain-of-thought reasoning +1
  4. CUJBench: Benchmarking LLM-Agent on Cross-Modal Failure Diagnosis from Browser to BackendLLM agents and tool use → Software engineering coding agents
  5. When the Agent Is the Adversary: Architectural Requirements for Agentic AI Containment After the April 2026 Frontier Model EscapeLLM agents and tool use → Agent threats, guardrails and governance
  6. Beyond Local vs. External: A Game-Theoretic Framework for Trustworthy Knowledge AcquisitionLLM reasoning → Reasoning for agents, planning and code
  7. PushupBench: Your VLM is not good at counting pushupsMultimodal LLMs / vision-language models → VLM benchmarks and visual reasoning
  8. Breaking the Resource Wall: Geometry-Guided Sequence Modeling for Efficient Semantic SegmentationState-space sequence models (S4, Mamba) → Vision Mamba backbones
  9. SoccerRef-Agents: Multi-Agent System for Automated Soccer RefereeingLLM agents and tool use → Multimodal and embodied agents
  10. V-GRPO: Online Reinforcement Learning for Denoising Generative Models Is Easier than You ThinkLLM reasoning → Reinforcement learning for reasoning
  11. Ghost in the Agent: Redefining Information Flow Tracking for LLM AgentsLLM agents and tool use → Agent threats, guardrails and governance
  12. GSAR: Typed Grounding for Hallucination Detection and Recovery in Multi-Agent LLMsLLM agents and tool use → Medical and scientific domain agents
  13. UniAda: Universal Adaptive Multi-objective Adversarial Attack for End-to-End Autonomous Driving SystemsEnd-to-end autonomous driving → Adversarial scenarios and safety testing
  14. Learning from Demonstration with Failure Awareness for Safe Robot NavigationOffline RL and imitation learning from data → Language-conditioned policies and VLAs
  15. When Chain-of-Thought Fails, the Solution Hides in the Hidden StatesLLM reasoning → Prompting and chain-of-thought reasoning
  16. GeoFunFlow-3D: A Physics-Guided Generative Flow Matching Framework for High-Fidelity 3D Aerodynamic Inference over Complex GeometriesDiffusion models → Flow matching, time series and science
  17. EmoTrans: A Benchmark for Understanding, Reasoning, and Predicting Emotion Transitions in Multimodal LLMsLLM reasoning → Audio-language model reasoning +1
  18. Bridging Reasoning and Action: Hybrid LLM-RL Framework for Efficient Cross-Domain Task-Oriented DialogueLLM reasoning → Reinforcement learning for reasoning
  19. Evaluating Jailbreaking Vulnerabilities in LLMs Deployed as Assistants for Smart Grid Operations: A Benchmark Against NERC StandardsLLM safety and security → Jailbreaking
  20. A Systematic Survey of Security Threats and Defenses in LLM-Based AI Agents: A Layered Attack Surface FrameworkLLM agents and tool use → Agent threats, guardrails and governance +1
  21. Efficient Rationale-based Retrieval: On-policy Distillation from Generative Rerankers based on JEPAKnowledge distillation → Knowledge distillation for LLMs
  22. Process Supervision of Confidence Margin for Calibrated LLM ReasoningLLM reasoning → Reinforcement learning for reasoning
  23. Hidden States Know Where Reasoning Diverges: Credit Assignment via Span-Level Wasserstein DistanceLLM reasoning → Reinforcement learning for reasoning +1
  24. GIFT: Global stabilisation via Intrinsic Fine TuningDeep reinforcement learning methods → RL theory and sample complexity
  25. CODA: Coordination via On-Policy Diffusion for Multi-Agent Offline Reinforcement LearningOffline RL and imitation learning from data → Offline multi-agent RL
  26. Towards Agentic Test-Driven Quality Assurance for 6G NetworksLLM agents and tool use → Software engineering coding agents
  27. Revisable by Design: A Theory of Streaming LLM Agent ExecutionLLM agents and tool use → General LLM agent frameworks and benchmarks
  28. Active Inference: A method for Phenotyping Agency in AI systems?LLM agents and tool use → General LLM agent frameworks and benchmarks
  29. From Similarity to Structure: Training-free LLM Context Compression with Hybrid Graph PriorsEfficient LLM inference and serving → Prompt and context compression
  30. Lightweight and Production-Ready PDF Visual Element ParsingRetrieval-augmented generation → Multimodal RAG
  31. Modular Sensory Stream for Integrating Physical Feedback in Vision-Language-Action ModelsVision-language-action models → Tactile, force-aware and humanoid VLA
  32. CAP-CoT: Cycle Adversarial Prompt for Improving Chain of Thoughts in LLM ReasoningLLM reasoning → Prompting and chain-of-thought reasoning +1
  33. MotionHiFlow: Text-to-motion via hierarchical flow matchingDiffusion models → Video generation
  34. AdaMamba: Adaptive Frequency-Gated Mamba for Long-Term Time Series ForecastingState-space sequence models (S4, Mamba) → Time series forecasting with SSMs
  35. Hiding in Plain Sight: Detectability-Aware Antidistillation of Reasoning ModelsLLM reasoning → Prompting and chain-of-thought reasoning +1
  36. Measuring Temporal Linguistic Emergence in Diffusion Language ModelsDiffusion models → Discrete and text diffusion
  37. Discovering Agentic Safety Specifications from 1-Bit Danger SignalsLLM agents and tool use → Agent threats, guardrails and governance
  38. From Coarse to Fine: Self-Adaptive Hierarchical Planning for LLM AgentsLLM agents and tool use → General LLM agent frameworks and benchmarks
  39. Scaling Multi-Node Mixture-of-Experts Inference Using Expert Activation PatternsMixture-of-experts → MoE large language models
  40. UpstreamQA: A Modular Framework for Explicit Reasoning on Video Question Answering TasksLLM reasoning → Multimodal reasoning
  41. UNSEEN: A Cross-Stack LLM Unlearning Defense against AR-LLM Social Engineering AttacksLLM safety and security → Prompt injection and agent security
  42. How Researchers Navigate Accountability, Transparency, and Trust When Using AI Tools in Early-Stage Research: A Think-Aloud StudyLLM reasoning → Knowledge- and tool-augmented reasoning
  43. From Concept-Aligned Tokens to Vulnerable Features: Mechanistic Localization of JailbreaksLLM safety and security → Jailbreaking +1
  44. ArgRE: Formal Argumentation for Conflict Resolution in Multi-Agent Requirements NegotiationLLM agents and tool use → Software engineering coding agents
  45. Breaking Lock-In: Preserving Steerability under Low-Data VLA Post-TrainingVision-language-action models → VLA architectures and generalization
  46. Mixture of Heterogeneous Grouped Experts for Language ModelingMixture-of-experts → Efficient MoE training and inference systems
  47. No Test Cases, No Problem: Distillation-Driven Code Generation for Scientific WorkflowsLLM agents and tool use → Software engineering coding agents
  48. Towards Automated Ontology Generation from Unstructured Text: A Multi-Agent LLM ApproachLLM agents and tool use → Web, search, and research agents
  49. Code Broker: A Multi-Agent System for Automated Code Quality AssessmentLLM agents and tool use → Software engineering coding agents
  50. Usable Agent Discovery for Decentralized AI SystemsLLM agents and tool use → General LLM agent frameworks and benchmarks
  51. From Pixels to Explanations: Interpretable Diabetic Retinopathy Grading with CNN-Transformer Ensembles, Visual Explainability and Vision-Language ModelsMultimodal LLMs / vision-language models → Medical vision-language models
  52. Benchmarking Storage Systems for Machine Learning Workloads Using NIO Bench
  53. Towards an Argumentative Foundation for Evaluative AI
  54. IdiomX A Multilingual Benchmark for Idiom Understanding, Retrieval, and Interpretation
  55. Fuzzy, Neutrosophic, and Uncertain Graph Theory: Properties and Applications
  56. Robustness of Refugee-Matching Gains to Off-Policy Evaluation Choices
  57. From Canopy to Collision: A Hybrid Predictive Framework for Identifying Risk Factors in Tree-Involved Traffic Crashes
  58. OceanPile: A Large-Scale Multimodal Ocean Corpus for Foundation Models
  59. Visual Chart Representations for Cryptocurrency Regime Prediction: A Systematic Deep Learning Study
  60. Latent Space Probing for Adult Content Detection in Video Generative Models
  61. V.O.I.C.E (Voice, Ownership, Identity, Control, Expression): Risk Taxonomy of Synthetic Voice Generation From Empirical Data
  62. CRC-SAM: SAM-Based Multi-Modal Segmentation and Quantification of Colorectal Cancer in CT, Colonoscopy, and Histology Images
  63. Semantic Denial of Service in LLM-controlled robots
  64. GeoCert: Certified Geometric AI for Reliable Forecasting
  65. Escher-Loop: Mutual Evolution by Closed-Loop Self-Referential Optimization
  66. Can Humans Detect AI? Mining Textual Signals of AI-Assisted Writing Under Varying Scrutiny Conditions
  67. Progress in Formalizing Sphere Packing in Dimension 8
  68. Hybrid JIT-CUDA Graph Optimization for Low-Latency Large Language Model Inference
  69. Evaluating CUDA Tile for AI Workloads on Hopper and Blackwell GPUs
  70. Machine learning models for estimating counterfactuals in a single-arm inflammatory bowel disease study
  71. Architecture Matters for Multi-Agent Security
  72. A Benchmark Suite of Reddit-Derived Datasets for Mental Health Detection
  73. Anchored Variational Inference for Personalized Sequential Latent-State Models
  74. From Edges to Depth: Probing the Spatial Hierarchy in Vision Transformers
  75. ArguAgent: AI-Supported Real-Time Grouping for Productive Argumentation in STEM Classrooms
  76. IndustryAssetEQA: A Neurosymbolic Operational Intelligence System for Embodied Question Answering in Industrial Asset Maintenance
  77. AI Safety Training Can be Clinically Harmful
  78. Revisiting Greedy Decoding for Visual Question Answering: A Calibration Perspective
  79. Resource-Constrained UAV-Based Weed Detection for Site-Specific Management on Edge Devices
  80. Scalable and Verifiable Federated Learning for Cross-Institution Financial Fraud Detection
  81. Inference of Online Newton Methods with Nesterov's Accelerated Sketching
  82. Knee-xRAI: An Explainable AI Framework for Automatic Kellgren-Lawrence Grading of Knee Osteoarthritis
  83. When Does Removing LayerNorm Help? Activation Bounding as a Regime-Dependent Implicit Regularizer
  84. Sphere-Depth: A Benchmark for Depth Estimation Methods with Varying Spherical Camera Orientations
  85. Automating Categorization of Scientific Texts with In-Context Learning and Prompt-Chaining in Large Language Models
  86. On (not) learning the M"obius function
  87. Enhanced Privacy and Communication Efficiency in Non-IID Federated Learning with Adaptive Quantization and Differential Privacy
  88. Evolve: A Persistent Knowledge Lifecycle for Small Language Models
  89. Approximating Uniform Random Rotations by Two-Stage Structured Hadamard Rotations in High Dimensions
  90. A Heterogeneous Two-Stream Framework for Video Action Recognition with Comparative Fusion Analysis
  91. Overcoming Copyright Barriers in Corpus Distribution Through Non-Reversible Hashing
  92. Learn&Drop: Fast Learning of CNNs based on Layer Dropping
  93. When Corrective Hints Hurt: Prompt Design in Reasoner-Guided Repair of LLM Overcaution on Entailed Negations under OWL~2~DL
  94. Lost in Decoding? Reproducing and Stress-Testing the Look-Ahead Prior in Generative Retrieval
  95. A Parametric Memory Head for Continual Generative Retrieval
  96. Keypoint-based Dynamic Object 6-DoF Pose Tracking via Event Camera
  97. A Taxonomy and Resolution Strategy for Client-Level Disagreements in Federated Learning
  98. Domain-Adapted Fine-Tuning of ECG Foundation Models for Multi-Label Structural Heart Disease Screening
  99. MCMC with Adaptive Principal-Component Transformation: Rotation-Invariant Universal Samplers for Bayesian Structural System Identification
  100. Constraint-Based Analysis of Reasoning Shortcuts in Neurosymbolic Learning
All 178
  1. Hierarchical Spatio-Channel Clustering for Efficient Model Compression in Medical Image Analysis
  2. Physics-Informed Temporal U-Net for High-Fidelity Fluid Interpolation
  3. When Context Sticks: Studying Interference in In-Context Learning
  4. Nonlinear Non-Gaussian Density Steering with Input and Noise Channel Mismatch: Sinkhorn with Memory for Solving the Control-affine Schr"odinger Bridge Problem
  5. TEMPO: Transformers for Temporal Disease Progression from Cross-Sectional Data
  6. An Empirical Evaluation of Locally Deployed LLMs for Bug Detection in Python Code
  7. VeriLLMed: Interactive Visual Debugging of Medical Large Language Models with Knowledge Graphs
  8. LEGO: An LLM Skill-Based Front-End Design Generation Platform
  9. Explainable AI in Speaker Recognition -- Making Latent Representations Understandable
  10. Evaluating Large Language Models on Computer Science University Exams in Data Structures
  11. Exploring Hierarchical Consistency and Unbiased Objectness for Open-Vocabulary Object Detection
  12. H-SemiS: Hierarchical Fusion of Semi and Self-Supervised Learning for Knee Osteoarthritis Severity Grading
  13. EAD-Net: Emotion-Aware Talking Head Generation with Spatial Refinement and Temporal Coherence
  14. Layer Embedding Deep Fusion Graph Neural Network
  15. Robust Audio-Text Retrieval via Cross-Modal Attention and Hybrid Loss
  16. KAConvNet: Kolmogorov-Arnold Convolutional Networks for Vision Recognition
  17. Learning from Noisy Prompts: Saliency-Guided Prompt Distillation for Robust Segmentation with SAM
  18. STAND: Semantic Anchoring Constraint with Dual-Granularity Disambiguation for Remote Sensing Image Change Captioning
  19. CombiMOTS: Combinatorial Multi-Objective Tree Search for Dual-Target Molecule Generation
  20. S2IT: Stepwise Syntax Integration Tuning for Large Language Models in Aspect Sentiment Quad Prediction
  21. Human-1 by Josh Talks: A Full-Duplex Conversational Modeling Framework in Hindi using Real-World Conversations
  22. An Analysis of Active Learning Algorithms using Real-World Crowd-sourced Text Annotations
  23. MetaErr: Towards Predicting Error Patterns in Deep Neural Networks
  24. Au-M-ol: A Unified Model for Medical Audio and Language Understanding
  25. Bridging the Pose-Semantic Gap: A Cascade Framework for Text-Based Person Anomaly Search
  26. Contrastive Learning for Multimodal Human Activity Recognition with Limited Labeled Data
  27. AI Identity: Standards, Gaps, and Research Directions for AI Agents
  28. SemiGDA: Generative Dual-distribution Alignment for Semi-Supervised Medical Image Segmentation
  29. A Hierarchical Ensemble Inference Pipeline for Robust White Blood Cell Classification Under Domain Shifts
  30. LatentBurst: A Fast and Efficient Multi Frame Super-Resolution for Hexadeca-Bayer Pattern CIS images
  31. Fine-tuning vs. In-context Learning in Large Language Models: A Formal Language Learning Perspective
  32. Small Language Model Helps Resolve Semantic Ambiguity of LLM Prompt
  33. Explicit integral representations and quantitative bounds for two-layer ReLU networks
  34. Knowledge Lever Risk Management for Software Engineering: A Stochastic Framework for Mitigating Knowledge Loss
  35. Architecture-Induced Recoverability Bias in Differentiable Symbolic Regression
  36. Scalable LLM-based Coding of Dialogue in Healthcare Simulation: Balancing Coding Performance, Processing Time, and Environmental Impact
  37. AI-Assisted Code Review as a Scaffold for Code Quality and Self-Regulated Learning: An Experience Report
  38. Micro-Expression-Aware Avatar Fingerprinting via Inter-Frame Feature Differencing
  39. Training Machine Learning Models on Encrypted Data: A Privacy-Preserving Framework using Homomorphic Encryption
  40. Spectro-Temporal Modulation Representation Framework for Human-Imitated Speech Detection
  41. Toward Polymorphic Backdoor against Semantic Communication via Intensity-Based Poisoning
  42. A Layer Separation Optimization Framework for Cross-Entropy Training in Deep Learning
  43. DARC-CLIP: Dynamic Adaptive Refinement with Cross-Attention for Meme Understanding
  44. Learning Curves and Benign Overfitting of Spectral Algorithms in Large Dimensions
  45. Tessera: Secure, Near-Line-Rate Weight Streaming for UMA Edge Accelerators
  46. StoryTR: Narrative-Centric Video Temporal Retrieval with Theory of Mind Reasoning
  47. Follow the TRACE: Exploiting Post-Click Trajectories for Online Delayed Conversion Rate Prediction
  48. AnalogRetriever: Learning Cross-Modal Representations for Analog Circuit Retrieval
  49. Well-Conditioned Oblivious Perturbations in Linear Space
  50. RAT: RunAnyThing via Fully Automated Environment Configuration
  51. DyABD: The Abdominal Muscle Segmentation in Dynamic MRI Benchmark
  52. A Unified Fractional Regularization Framework for Sparse Recovery
  53. Designing escalation criteria for international AI incident response: criteria, triggers, and thresholds
  54. Cooperative Informative Sensing for Monitoring Dynamic Indoor Environments via Multi-Agent Reinforcement Learning
  55. Judging the Judges: A Systematic Evaluation of Bias Mitigation Strategies in LLM-as-a-Judge Pipelines
  56. One Identity, Many Roles: Multimodal Entity Coreference for Enhanced Video Situation Recognition
  57. Efficient VQ-QAT and Mixed Vector/Linear quantized Neural Networks
  58. A Topology fixated Shape Gradient Framework for Non Simple Boundary Extraction for CIE Lab color images with Repulsive Energy
  59. A satellite foundation model for improved wealth monitoring
  60. BSViT: A Burst Spiking Vision Transformer for Expressive and Efficient Visual Representation Learning
  61. PhySE: A Psychological Framework for Real-Time AR-LLM Social Engineering Attacks
  62. CNN-ViT Fusion with Adaptive Attention Gate for Brain Tumor MRI Classification: A Hybrid Deep Learning Model
  63. Characterizing Paraphrase-Induced Failures in Lean 4 Autoformalization
  64. h-MINT: Modeling Pocket-Ligand Binding with Hierarchical Molecular Interaction Network
  65. MindTrellis: Co-Creating Knowledge Structures with AI through Interactive Visual Exploration
  66. A Dynamic Learning Observatory Reveals the Rapid Salinization of Satkhira, Bangladesh
  67. Learning from Imperfect Text Guidance: Robust Long-Tail Visual Recognition with High-Noise Label
  68. HBGSA: Hydrogen Bond Graph with Self-Attention for Drug-Target Binding Affinity Prediction
  69. A Tale of Two Variances: When Single-Seed Benchmarks Fail in Bayesian Deep Learning
  70. Conditional Imputation for Within-Modality Missingness in Multi-Modal Federated Learning
  71. MOCA: A Transformer-based Modular Causal Inference Framework with One-way Cross-attention and Cutting Feedback
  72. Transferable Physical-World Adversarial Patches Against Object Detection in Autonomous Driving
  73. Unstable Rankings in Bayesian Deep Learning Evaluation
  74. ProEval: Proactive Failure Discovery and Efficient Performance Estimation for Generative AI Evaluation
  75. INSIGHT: Indoor Scene Intelligence from Geometric-Semantic Hierarchy Transfer for Public~Safety
  76. FusionRelight: Relighting Portraits in Real Time via Hybrid Domain Knowledge Fusion
  77. Channel Adaptation for EEG Foundation Models: A Systematic Benchmark Across Architectures, Tasks, and Training Regimes
  78. Turtle shell clustering: A mixture approach to discriminative clustering with applications to flow cytometry and other data