mlsanityData to 14 Oct 2026

Papers

New on arXiv

Sun 2 Feb 2025

174 papers

  1. Safety at Scale: A Comprehensive Survey of Large Model and Agent SafetyMultimodal LLMs / vision-language models → MLLM jailbreaks and adversarial robustness +2
  2. Metastable Dynamics of Chain-of-Thought Reasoning: Provable Benefits of Search, RL and DistillationLLM reasoning → Prompting and chain-of-thought reasoning +1
  3. Fast Direct: Query-Efficient Online Black-box Guidance for Diffusion-model Target GenerationDiffusion models → Diffusion foundations and sampling
  4. Agent-Based Uncertainty Awareness Improves Automated Radiology Report Labeling with an Open-Source Large Language ModelLLM agents and tool use → Medical and scientific domain agents
  5. HuViDPO:Enhancing Video Generation through Direct Preference Optimization for Human-Centric AlignmentDiffusion models → Video generation +1
  6. Efficient Multi-Agent System Training with Data Influence-Oriented Tree SearchLLM agents and tool use → General LLM agent frameworks and benchmarks +1
  7. Hypo3D: Exploring Hypothetical Reasoning in 3DMultimodal LLMs / vision-language models → Embodied and 3D vision-language models
  8. ModServe: Modality- and Stage-Aware Resource Disaggregation for Scalable Multimodal Model ServingMultimodal LLMs / vision-language models → Visual token pruning and efficiency
  9. Generalizing Safety Beyond Collision-Avoidance via Latent-Space Reachability AnalysisWorld models → Latent dynamics models and model-based RL
  10. VL-Nav: Neuro-Symbolic Reasoning-based Vision-Language NavigationLLM reasoning → Spatial and embodied reasoning
  11. Huff-LLM: End-to-End Lossless Compression for Efficient LLM InferenceEfficient LLM inference and serving → KV cache optimization
  12. Blink of an eye: a simple theory for feature localization in generative modelsDiffusion models → Discrete and text diffusion
  13. SimPER: A Minimalist Approach to Preference Alignment without HyperparametersLLM alignment / RLHF / preference optimisation → LLM preference optimization methods
  14. Language Models Use Trigonometry to Do AdditionLLM reasoning → Mathematical reasoning
  15. FedHPD: Heterogeneous Federated Reinforcement Learning via Policy DistillationDeep reinforcement learning methods → RL for energy, networks and operations
  16. Inferring the Unspoken: Aligning Embodied Agents with Implicit PreferencesLLM agents and tool use → Multimodal and embodied agents +1
  17. Dual Alignment Maximin Optimization for Offline Model-based RLOffline RL and imitation learning from data → Offline RL algorithms
  18. RealRAG: Retrieval-augmented Realistic Image Generation via Self-reflective Contrastive LearningRetrieval-augmented generation → Multimodal RAG +1
  19. VLM-Assisted Continual learning for Visual Question Answering in Self-DrivingMultimodal LLMs / vision-language models → Driving VLMs and open-vocabulary detection +2
  20. CAIMAN: Causal Action Influence Detection for Sample-efficient Loco-manipulationDeep reinforcement learning methods → RL for robot locomotion and manipulation
  21. Weak Supervision Dynamic KL-Weighted Diffusion Models Guided by Large Language ModelsDiffusion models → Image generation, text-to-image and editing
  22. Probing Large Language Models in Reasoning and Translating Complex Linguistic PuzzlesLLM reasoning → Prompting and chain-of-thought reasoning
  23. Disentangling Length Bias In Preference Learning Via Response-Conditioned ModelingLLM alignment / RLHF / preference optimisation → Reward modeling and RLHF
  24. Synthetic Artifact Auditing: Tracing LLM-Generated Synthetic Data Usage in Downstream ApplicationsSynthetic data and self-improvement for LLMs → LLM synthetic data generation
  25. UniGraph2: Learning a Unified Embedding Space to Bind Multimodal GraphsMixture-of-experts → Multimodal MoE models
  26. Fisher-Guided Selective Forgetting: Mitigating The Primacy Bias in Deep Reinforcement LearningDeep reinforcement learning methods → Deep RL algorithms and representations
  27. RTBAgent: A LLM-based Agent System for Real-Time BiddingLLM agents and tool use → Financial trading and forecasting agents
  28. Vision-centric Token Compression in Large Language ModelEfficient LLM inference and serving → Visual token pruning in multimodal models
  29. Role of Mixup in Topological Persistence Based Knowledge Distillation for Wearable Sensor DataKnowledge distillation → General knowledge distillation methods
  30. AgentBreeder: Mitigating the AI Safety Risks of Multi-Agent Scaffolds via Self-ImprovementLLM agents and tool use → General LLM agent frameworks and benchmarks +1
  31. Universal Post-Processing Networks for Joint Optimization of Modules in Task-Oriented Dialogue SystemsDeep reinforcement learning methods → RL training of language and agent models
  32. BEEM: Boosting Performance of Early Exit DNNs using Multi-Exit Classifiers as ExpertsEfficient LLM inference and serving → Early exit and edge inference
  33. `From Prompt to Perturbation': An Adaptive Framework for Voice-Based Jailbreaks on Audio LLMsLLM safety and security → Audio jailbreaks and speech backdoors
  34. Perspectives for Direct Interpretability in Multi-Agent Deep Reinforcement LearningMechanistic interpretability → Circuits and mechanistic analysis
  35. Demystifying Cost-Efficiency in LLM Serving over Heterogeneous GPUsEfficient LLM inference and serving → Serving systems and request scheduling
  36. "I am bad": Interpreting Stealthy, Universal and Robust Audio Jailbreaks in Audio-Language ModelsLLM safety and security → Audio jailbreaks and speech backdoors
  37. MINT: Mitigating Hallucinations in Large Vision-Language Models via Token ReductionMultimodal LLMs / vision-language models → Hallucination mitigation in LVLMs +1
  38. VIKSER: Visual Knowledge-Driven Self-Reinforcing Reasoning FrameworkLLM reasoning → Multimodal reasoning
  39. PhiP-G: Physics-Guided Text-to-3D Compositional Scene GenerationWorld models → Video-generation world models and simulators +1
  40. To Code or not to Code? Adaptive Tool Integration for Math Language Models via Expectation-MaximizationLLM reasoning → Reasoning for agents, planning and code
  41. High-Order Matching for One-Step Shortcut Diffusion ModelsDiffusion models → Flow matching, time series and science
  42. Compositional Concept-Based Neuron-Level Interpretability for Deep Reinforcement LearningDeep reinforcement learning methods → Interpretable and logic-guided RL
  43. ReFoRCE: A Text-to-SQL Agent with Self-Refinement, Consensus Enforcement, and Column ExplorationLLM agents and tool use → Web, search, and research agents
  44. Rethinking Mixture-of-Agents: Is Mixing Different Large Language Models Beneficial?LLM reasoning → Test-time compute and efficient thinking +1
  45. Safety Alignment Depth in Large Language Models: A Markov Chain PerspectiveLLM alignment / RLHF / preference optimisation → Safety alignment and jailbreak attacks +1
  46. Avoiding mathbfexp(Rmax) scaling in RLHF through Preference-based ExplorationLLM alignment / RLHF / preference optimisation → Reward modeling and RLHF
  47. Mitigating the Modality Gap: Few-Shot Out-of-Distribution Detection with Multi-modal Prototypes and Image Bias EstimationMultimodal LLMs / vision-language models → Prompt learning and adaptation of CLIP
  48. LLM Safety Alignment is Divergence Estimation in DisguiseLLM alignment / RLHF / preference optimisation → Safety alignment and jailbreak attacks +1
  49. Towards Robust Multimodal Large Language Models Against Jailbreak AttacksLLM safety and security → Vision-language and text-to-image safety
  50. Reformulation is All You Need: Addressing Malicious Text Features in DNNsLLM safety and security → Backdoor and data poisoning attacks
  51. Half-order Fine-Tuning for Diffusion Model: A Recursive Likelihood Ratio OptimizerDiffusion models → Diffusion and flow policies for robotics
  52. SimulPL: Aligning Human Preferences in Simultaneous Machine TranslationLLM alignment / RLHF / preference optimisation → LLM preference optimization methods
  53. Advanced Weakly-Supervised Formula Exploration for Neuro-Symbolic Mathematical ReasoningLLM reasoning → Reasoning benchmarks and evaluation
  54. Inference-Time Enhancement of Generative Robot Policies via Predictive World ModelingWorld models → Embodied and robot world models +1
  55. DesCLIP: Robust Continual Learning via General Attribute Descriptions for VLM-Based Visual RecognitionMultimodal LLMs / vision-language models → Prompt learning and adaptation of CLIP
  56. Enhancing Code Consistency in AI Research with Large Language Models and Retrieval-Augmented GenerationRetrieval-augmented generation → RAG for code
  57. Enhancing Offline Reinforcement Learning with Curriculum Learning-Based Trajectory ValuationOffline RL and imitation learning from data → Offline RL algorithms
  58. What can large language models do for sustainable food?
  59. "Would You Want an AI Tutor?" Understanding Stakeholder Perceptions of LLM-based Systems in the Classroom
  60. Bridging Simulation and Reality: A 3D Clustering-Based Deep Learning Model for UAV-Based RF Source Localization
  61. neuro2voc: Decoding Vocalizations from Neural Activity
  62. Emotion Recognition and Generation: A Comprehensive Review of Face, Speech, and Text Modalities
  63. A Novel Real-Time Full-Color 3D Holographic (Diffractive) Video Capture, Processing, and Transmission Pipeline Using Off-The-Shelf Hardware
  64. Predicting Steady-State Behavior in Complex Networks with Graph Neural Networks
  65. scGSDR: Harnessing Gene Semantics for Single-Cell Pharmacological Profiling
  66. BrainOOD: Out-of-distribution Generalizable Brain Network Analysis
  67. Automated Extraction of Spatio-Semantic Graphs for Identifying Cognitive Impairment
  68. Predict, Cluster, Refine: A Joint Embedding Predictive Self-Supervised Framework for Graph Representation Learning
  69. LLM-Powered Benchmark Factory: Reliable, Generic, and Efficient
  70. The exception of humour: Iconicity, Phonemic Surprisal, Memory Recall, and Emotional Associations
  71. DeepGate4: Efficient and Effective Representation Learning for Circuit Design at Scale
  72. Neurosymbolic AI for Travel Demand Prediction: Integrating Decision Tree Rules into Neural Networks
  73. LIBRA: Measuring Bias of Large Language Model from a Local Context
  74. LEAD: An EEG Foundation Model for Alzheimer's Disease Detection
  75. Position: AI Scaling: From Up to Down and Out
  76. SAM-guided Pseudo Label Enhancement for Multi-modal 3D Semantic Segmentation
  77. Minimax Optimality of Classical Scaling Under General Noise Conditions
  78. Training speedups via batching for geometric learning: an analysis of static and dynamic algorithms
  79. Universal Abstraction: Harnessing Frontier Models to Structure Real-World Data at Scale
  80. An MDP Model for Censoring in Harvesting Sensors: Optimal and Approximated Solutions
  81. Fruit Fly Classification (Diptera: Tephritidae) in Images, Applying Transfer Learning
  82. Attention Sinks: A 'Catch, Tag, Release' Mechanism for Embeddings
  83. The Accuracy, Robustness, and Readability of LLM-Generated Sustainability-Related Word Definitions
  84. Embracing Dialectic Intersubjectivity: Coordination of Different Perspectives in Content Analysis with LLM Persona Simulation
  85. More Rigorous Software Engineering Would Improve Reproducibility in Machine Learning Research
  86. Fundamental limits of learning in sequence multi-index models and deep attention networks: High-dimensional asymptotics and sharp thresholds
  87. HASSLE-free: A unified Framework for Sparse plus Low-Rank Matrix Decomposition for LLMs
  88. Multi-frequency wavefield solutions for variable velocity models using meta-learning enhanced low-rank physics-informed neural network
  89. LoR-VP: Low-Rank Visual Prompting for Efficient Vision Model Adaptation
  90. MorphBPE: A Morpho-Aware Tokenizer Bridging Linguistic Complexity for Efficient LLM Training Across Morphologies
  91. Algorithmic Stability of Stochastic Gradient Descent with Momentum under Heavy-Tailed Noise
  92. Worth Their Weight: Randomized and Regularized Block Kaczmarz Algorithms without Preprocessing
  93. Generating Computational Cognitive Models using Large Language Models
  94. Position: The Artificial Intelligence and Machine Learning Community Should Adopt a More Transparent and Regulated Peer Review Process
  95. Modified Adaptive Tree-Structured Parzen Estimator for Hyperparameter Optimization
  96. A Unified Theory of Sinusoidal Activation Families for Implicit Neural Representations
  97. Predicting potentially abusive clauses in Chilean terms of services with natural language processing
  98. FedRIR: Rethinking Information Representation in Federated Learning
  99. HintEval: An Open-Source Python Toolkit for Hint Generation and Hint Evaluation
  100. Psychometric-Based Evaluation for Theorem Proving with Large Language Models
All 174
  1. High-Dimensional Bayesian Optimization Using Both Random and Supervised Embeddings
  2. SecPE: Secure Prompt Ensembling for Private and Robust Large Language Models
  3. Federated Generalised Variational Inference: A Robust Probabilistic Federated Learning Framework
  4. Activation Approximations Can Incur Safety Vulnerabilities Even in Aligned LLMs: Comprehensive Analysis and Defense
  5. Explainability in Practice: A Survey of Explainable NLP Across Various Domains
  6. Boosting Adversarial Robustness and Generalization with Dictionary Structure
  7. Cross multiscale vision transformer for deep fake detection
  8. Generalization of Medical Large Language Models through Cross-Domain Weak Supervision
  9. When Do LLMs Help With Node Classification? A Comprehensive Analysis
  10. Decision-informed Neural Networks with Large Language Model Integration for Portfolio Optimization
  11. Online Learning of Pure States is as Hard as Mixed States
  12. OOD Detection with immature Models
  13. Error-quantified Conformal Inference for Time Series
  14. Sundial: A Family of Highly Capable Time Series Foundation Models
  15. ProPINN: Demystifying Propagation Failures in Physics-Informed Neural Networks
  16. Environment-Driven Online LiDAR-Camera Extrinsic Calibration
  17. Adversarial Semantic Augmentation for Training Generative Adversarial Networks under Limited Data
  18. Deep Neural Network for Phonon-Assisted Optical Spectra in Semiconductors
  19. Task-Specific Adaptation with Restricted Model Access
  20. Estimating forest carbon stocks from high-resolution remote sensing imagery by reducing domain shift with style transfer
  21. A method for estimating forest carbon storage distribution density via artificial intelligence generated content model
  22. Transfer Learning in Physics-Informed Neural Networks: Full Fine-Tuning, Lightweight Fine-Tuning, and Low-Rank Adaptation
  23. ATA: Adaptive Task Allocation for Efficient Resource Management in Distributed Machine Learning
  24. Learning-Based TSP-Solvers Tend to Be Overly Greedy
  25. FIRE: Flexible Integration of Data Quality Ratings for Effective Pre-Training
  26. Privacy Preserving Properties of Vision Classifiers
  27. Structural Latency Perturbation in Large Language Models Through Recursive State Induction
  28. Continuity-Preserving Convolutional Autoencoders for Learning Continuous Latent Dynamical Models from Images
  29. Mirror Descent Under Generalized Smoothness
  30. Zero-Shot Warning Generation for Misinformative Multimodal Content
  31. An Event-Based Perception Pipeline for a Table Tennis Robot
  32. CoNNect: Connectivity-Based Regularization for Structural Pruning
  33. An Efficient Orlicz-Sobolev Approach for Transporting Unbalanced Measures on a Graph
  34. Scalable Sobolev IPM for Probability Measures on a Graph
  35. CycleGuardian: A Framework for Automatic RespiratorySound classification Based on Improved Deep clustering and Contrastive Learning
  36. Spatio-Temporal Progressive Attention Model for EEG Classification in Rapid Serial Visual Presentation Task
  37. Selective Response Strategies for GenAI
  38. Meta-Prompt Optimization for LLM-Based Sequential Decision Making
  39. Understanding and Mitigating the High Computational Cost in Path Data Diffusion
  40. Learned Bayesian Cram'er-Rao Bound for Unknown Measurement Models Using Score Neural Networks
  41. Vision and Language Reference Prompt into SAM for Few-shot Segmentation
  42. UPL: Uncertainty-aware Pseudo-labeling for Imbalance Transductive Node Classification
  43. Registration-Enhanced Segmentation Method for Prostate Cancer in Ultrasound Images
  44. Model Provenance Testing for Large Language Models
  45. Optimization for Neural Operators can Benefit from Width
  46. S2CFormer: Revisiting the RD-Latency Trade-off in Transformer-based Learned Image Compression
  47. MM-IQ: Benchmarking Human-Like Abstraction and Reasoning in Multimodal Models
  48. TMI-CLNet: Triple-Modal Interaction Network for Chronic Liver Disease Prognosis From Imaging, Clinical, and Radiomic Data Fusion
  49. Leveraging Large Language Models to Predict Antibody Biological Activity Against Influenza A Hemagglutinin
  50. Dissecting Submission Limit in Desk-Rejections: A Mathematical Analysis of Fairness in AI Conference Policies
  51. Guidance Source Matters: How Guidance from AI, Expert, or a Group of Analysts Impacts Visual Data Preparation and Analysis
  52. A Survey of Quantized Graph Representation Learning: Connecting Graph Structures with Large Language Models
  53. How Contaminated Is Your Benchmark? Quantifying Dataset Leakage in Large Language Models with Kernel Divergence
  54. LLM-based event log analysis techniques: A survey
  55. Biogeochemistry-Informed Neural Network (BINN) for Improving Accuracy of Model Prediction and Scientific Understanding of Soil Organic Carbon
  56. Cross-Modal Synergies: Unveiling the Potential of Motion-Aware Fusion Networks in Handling Dynamic and Static ReID Scenarios
  57. Enhanced Convolutional Neural Networks for Improved Image Classification
  58. EmoTalkingGaussian: Continuous Emotion-conditioned Talking Head Synthesis
  59. Agency in the Age of AI
  60. TrojanTime: Backdoor Attacks on Time Series Classification
  61. General Coded Computing in a Probabilistic Straggler Regime
  62. Evaluating Small Language Models for News Summarization: Implications and Factors Influencing Performance
  63. CollabLLM: From Passive Responders to Active Collaborators
  64. Lipschitz Lifelong Monte Carlo Tree Search for Mastering Non-Stationary Tasks
  65. MedConv: Convolutions Beat Transformers on Long-Tailed Bone Density Prediction
  66. Self-Prompt SAM: Medical Image Segmentation via Automatic Prompt SAM Adaptation
  67. Representations Shape Weak-to-Strong Generalization: Theoretical Insights and Empirical Predictions
  68. Distribution-aware Fairness Learning in Medical Image Segmentation From A Control-Theoretic Perspective
  69. Efficient Language Modeling for Low-Resource Settings with Hybrid RNN-Transformer Architectures
  70. Using Causality for Enhanced Prediction of Web Traffic Time Series
  71. PAC Learning is just Bipartite Matching (Sort of)
  72. The Query/Hit Model for Sequential Hypothesis Testing
  73. Gradient Alignment in Physics-informed Neural Networks: A Second-Order Optimization Perspective
  74. Mitigating Heterogeneous Token Overfitting in LLM Knowledge Editing