mlsanityData to 14 Oct 2026

Papers

New on arXiv

Mon 5 May 2025

237 papers

  1. AKD : Adversarial Knowledge Distillation For Large Language Models Alignment on Coding tasksKnowledge distillation → Knowledge distillation for LLMs
  2. Learning 3D Persistent Embodied World ModelsWorld models → Video-generation world models and simulators
  3. Direct Retrieval-augmented Optimization: Synergizing Knowledge Selection and Language ModelsRetrieval-augmented generation → RAG pipeline design and evaluation
  4. Variational diffusion transformers for conditional sampling of supernovae spectraDiffusion models → Inverse problems and image restoration
  5. Improving Model Alignment Through Collective Intelligence of Open-Source LLMSLLM alignment / RLHF / preference optimisation → LLM preference optimization methods
  6. MORE: Mobile Manipulation Rearrangement Through Grounded Language ReasoningLLM reasoning → Spatial and embodied reasoning
  7. Adversarial Robustness Analysis of Vision-Language Models in Medical Image SegmentationMultimodal LLMs / vision-language models → Medical vision-language models
  8. A Fully Data-Driven Value Iteration for Stochastic LQR: Convergence, Robustness and StabilityDeep reinforcement learning methods → RL for games and combinatorial optimization
  9. RetroInfer: A Vector Storage Engine for Scalable Long-Context LLM InferenceEfficient LLM inference and serving → KV cache optimization
  10. Unlearning vs. Obfuscation: Are We Truly Removing Knowledge?LLM safety and security → Machine unlearning
  11. Rewriting Pre-Training Data Boosts LLM Performance in Math and CodeLLM reasoning → Reasoning for agents, planning and code
  12. Scenethesis: A Language and Vision Agentic Framework for 3D Scene GenerationLLM agents and tool use → Multimodal and embodied agents
  13. R1-Reward: Training Multimodal Reward Model Through Stable Reinforcement LearningMultimodal LLMs / vision-language models → Embodied and 3D vision-language models +1
  14. No Other Representation Component Is Needed: Diffusion Transformers Can Provide Representation Guidance by ThemselvesDiffusion models → Diffusion foundations and sampling
  15. LISAT: Language-Instructed Segmentation Assistant for Satellite ImageryMultimodal LLMs / vision-language models → Embodied and 3D vision-language models
  16. Towards Dataset Copyright Evasion Attack against Personalized Text-to-Image Diffusion ModelsDiffusion models → Robustness, privacy and adversarial aspects +1
  17. AutoLibra: Agent Metric Induction from Open-Ended Human FeedbackLLM agents and tool use → General LLM agent frameworks and benchmarks +1
  18. Knowing You Don't Know: Learning When to Continue Search in Multi-round RAG through Self-PracticingRetrieval-augmented generation → Conversational and speech RAG
  19. Advancing Generalizable Tumor Segmentation with Anomaly-Aware Open-Vocabulary Attention Maps and Frozen Foundation Diffusion ModelsDiffusion models → Medical imaging and segmentation
  20. FormalMATH: Benchmarking Formal Mathematical Reasoning of Large Language ModelsLLM reasoning → Mathematical reasoning
  21. Enhancing LLMs' Clinical Reasoning with Real-World Data from a Nationwide Sepsis RegistryLLM reasoning → Clinical and biomedical applications
  22. Graph Neural Network-Based Reinforcement Learning for Controlling Biological Networks - the GATTACA FrameworkDeep reinforcement learning methods → RL for energy, networks and operations
  23. Technical Report: Evaluating Goal Drift in Language Model AgentsLLM agents and tool use → General LLM agent frameworks and benchmarks
  24. Sailing by the Stars: A Survey on Reward Models and Learning Strategies for Learning from RewardsLLM reasoning → Reinforcement learning for reasoning +1
  25. A Survey on Progress in LLM Alignment from the Perspective of Reward DesignLLM alignment / RLHF / preference optimisation → Reward modeling and RLHF
  26. A Survey of Slow Thinking-based Reasoning LLMs using Reinforced Learning and Inference-time Scaling LawLLM reasoning → Test-time compute and efficient thinking
  27. MCCD: Multi-Agent Collaboration-based Compositional Diffusion for Complex Text-to-Image GenerationDiffusion models → Image generation, text-to-image and editing +1
  28. Transfer learning-enhanced deep reinforcement learning for aerodynamic airfoil optimisation subject to structural constraintsDeep reinforcement learning methods → RL with diffusion and flow models
  29. Detect, Classify, Act: Categorizing Industrial Anomalies with Multi-Modal Large Language ModelsMultimodal LLMs / vision-language models → Chart understanding and anomaly detection
  30. Neurodivergent Influenceability as a Contingent Solution to the AI Alignment ProblemLLM alignment / RLHF / preference optimisation → Human values and AI feedback alignment
  31. Recursive Decomposition with Dependencies for Generic Divide-and-Conquer ReasoningLLM reasoning → Test-time compute and efficient thinking
  32. Large Language Model Partitioning for Low-Latency Inference at the EdgeEfficient LLM inference and serving → Early exit and edge inference
  33. Resolving Memorization in Empirical Diffusion Model for Manifold Data in High-Dimensional SpacesDiffusion models → Diffusion foundations and sampling
  34. Beyond the model: Key differentiators in large language models and multi-agent servicesLLM agents and tool use → General LLM agent frameworks and benchmarks
  35. SEFE: Superficial and Essential Forgetting Eliminator for Multimodal Continual Instruction TuningMultimodal LLMs / vision-language models → Prompt learning and adaptation of CLIP
  36. El Agente: An Autonomous Agent for Quantum ChemistryLLM agents and tool use → Medical and scientific domain agents
  37. FairPO: Robust Preference Optimization for Fair Multi-Label LearningLLM alignment / RLHF / preference optimisation → LLM preference optimization methods
  38. SymbioticRAG: Enhancing Document Intelligence Through Human-LLM Symbiotic CollaborationRetrieval-augmented generation → RAG pipeline design and evaluation
  39. T2S: High-resolution Time Series Generation with Text-to-Series Diffusion ModelsDiffusion models → Flow matching, time series and science
  40. Task-Oriented Semantic Communication in Large Multimodal Models-based Vehicle NetworksMultimodal LLMs / vision-language models → Driving VLMs and open-vocabulary detection
  41. Optimizing Chain-of-Thought Reasoners via Gradient Variance Minimization in Rejection Sampling and RLLLM reasoning → Prompting and chain-of-thought reasoning
  42. RM-R1: Reward Modeling as ReasoningLLM reasoning → Reinforcement learning for reasoning +1
  43. EntroLLM: Entropy Encoded Weight Compression for Efficient Large Language Model Inference on Edge DevicesEfficient LLM inference and serving → KV cache optimization
  44. SIMPLEMIX: Frustratingly Simple Mixing of Off- and On-policy Data in Language Model Preference LearningLLM alignment / RLHF / preference optimisation → LLM preference optimization methods
  45. An End-to-End Model for Logits-Based Large Language Models WatermarkingLLM safety and security → Watermarking generated content and models
  46. TeDA: Boosting Vision-Lanuage Models for Zero-Shot 3D Object Retrieval via Testing-time Distribution AlignmentMultimodal LLMs / vision-language models → Embodied and 3D vision-language models
  47. HyperTree Planning: Enhancing LLM Reasoning via Hierarchical ThinkingLLM reasoning → Reasoning for agents, planning and code
  48. Perspective-Aware AI in Extended Reality
  49. Rapid yet accurate Tile-circuit and device modeling for Analog In-Memory Computing
  50. Predicting The Evolution of Interfaces with Fourier Neural Operators
  51. End-to-end fully-binarized network design: from Generic Learned Thermometer to Block Pruning
  52. Deep Learning Empowered Sub-Diffraction Terahertz Backpropagation Single-Pixel Imaging
  53. Moving From Monolithic To Microservices Architecture for Multi-Agent Systems
  54. ML-Enabled Eavesdropper Detection in Beyond 5G IIoT Networks
  55. Knowledge Guided Encoder-Decoder Framework: Integrating Multiple Physical Models for Agricultural Ecosystem Modeling
  56. ONERA's CRM WBPN database for machine learning activities, related regression challenge and first results
  57. Prediction of Delirium Risk in Mild Cognitive Impairment Using Time-Series data, Machine Learning and Comorbidity Patterns -- A Retrospective Study
  58. An Automated LLM-based Pipeline for Asset-Level Database Creation to Assess Deforestation Impact
  59. Computational Irreducibility as the Foundation of Agency: A Formal Model Connecting Undecidability to Autonomous Behavior in Complex Systems
  60. ChatGPT for automated grading of short answer questions in mechanical ventilation
  61. Prediction-powered estimators for finite population statistics in highly imbalanced textual data: Public hate crime estimation
  62. Rethinking Multimodal Sentiment Analysis: A High-Accuracy, Simplified Fusion Architecture
  63. A Comparative Benchmark of a Moroccan Darija Toxicity Detection Model (Typica.ai) and Major LLM-Based Moderation APIs (OpenAI, Mistral, Anthropic)
  64. Impact Analysis of Inference Time Attack of Perception Sensors on Autonomous Vehicles
  65. Improved Dimensionality Reduction for Inverse Problems in Nuclear Fusion and High-Energy Astrophysics
  66. Advanced Clustering Framework for Semiconductor Image Analytics Integrating Deep TDA with Self-Supervised and Transfer Learning Techniques
  67. GAME: Learning Multimodal Interactions via Graph Structures for Personality Trait Estimation
  68. A Deep Learning approach for Depressive Symptoms assessment in Parkinson's disease patients using facial videos
  69. From Spaceborne to Airborne: SAR Image Synthesis Using Foundation Models for Multi-Scale Adaptation
  70. Coverage Biases in High-Resolution Satellite Imagery
  71. CoCoB: Adaptive Collaborative Combinatorial Bandits for Online Recommendation
  72. Adaptive Data-Resilient Multi-Modal Hierarchical Multi-Label Book Genre Identification
  73. IntelliCardiac: An Intelligent Platform for Cardiac Image Segmentation and Classification
  74. Robustly Invertible Nonlinear Dynamics and the BiLipREN: Contracting Neural Models with Contracting Inverses
  75. BLAB: Brutally Long Audio Bench
  76. Developing A Framework to Support Human Evaluation of Bias in Generated Free Response Text
  77. Teaching Models to Understand (but not Generate) High-risk Data
  78. 34 Examples of LLM Applications in Materials Science and Chemistry: Towards Automation, Assistants, Agents, and Accelerated Scientific Discovery
  79. Sim2Real Transfer for Vision-Based Grasp Verification
  80. A New Perspective To Understanding Multi-resolution Hash Encoding For Neural Fields
  81. An Explainable Anomaly Detection Framework for Monitoring Depression and Anxiety Using Consumer Wearable Devices
  82. Dual Prompting for Diverse Count-level PET Denoising
  83. Modeling Spatial Extremes using Non-Gaussian Spatial Autoregressive Models via Convolutional Neural Networks
  84. Evaluating the Impact of AI-Powered Audiovisual Personalization on Learner Emotion, Focus, and Learning Outcomes
  85. Radio: Rate-Distortion Optimization for Large Language Model Compression
  86. UCSC at SemEval-2025 Task 3: Context, Models and Prompt Optimization for Automated Hallucination Detection in LLM Output
  87. A Typology of Synthetic Datasets for Dialogue Processing in Clinical Contexts
  88. The Multimodal Paradox: How Added and Missing Modalities Shape Bias and Performance in Multimodal AI
  89. Memorization or Interpolation ? Detecting LLM Memorization through Input Perturbation Analysis
  90. Lesion-Aware Generative Artificial Intelligence for Virtual Contrast-Enhanced Mammography in Breast Cancer
  91. GIF: Generative Inspiration for Face Recognition at Scale
  92. NTIRE 2025 Challenge on UGC Video Enhancement: Methods and Results
  93. RADLADS: Rapid Attention Distillation to Linear Attention Decoders at Scale
  94. New affine invariant ensemble samplers and their dimensional scaling
  95. More Optimal Fractional-Order Stochastic Gradient Descent for Non-Convex Optimization Problems
  96. Logits-Constrained Framework with RoBERTa for Ancient Chinese NER
  97. Completing Spatial Transcriptomics Data for Gene Expression Prediction Benchmarking
  98. Parameter estimation for land-surface models using Neural Physics
  99. PLAID: A Unified Data Model for Machine Learning on Heterogeneous Physics Simulations
  100. GeoERM: Geometry-Aware Multi-Task Representation Learning on Riemannian Manifolds
All 237
  1. Generating Narrated Lecture Videos from Slides with Synchronized Highlights
  2. Single-Sample and Robust Online Resource Allocation
  3. Smooth Quadratic Prediction Markets
  4. Iterative Resolution of Prompt Ambiguities Using a Progressive Cutting-Search Approach
  5. Gone With the Bits: Revealing Racial Bias in Low-Rate Neural Compression for Facial Images
  6. The Art of Repair: Optimizing Iterative Program Repair with Instruction-Tuned Models
  7. Early Prediction of Sepsis: Feature-Aligned Transfer Learning
  8. When Your Own Output Becomes Your Training Data: Noise-to-Meaning Loops and a Formal RSI Trigger
  9. CreoPep: A Universal Deep Learning Framework for Target-Specific Peptide Design and Optimization
  10. Taskmaster Deconstructed: A Quantitative Look at Tension, Volatility, and Viewer Ratings
  11. Beyond Fixed Patches: Enhancing GPTs for Financial Prediction with Adaptive Segmentation and Learnable Wavelets
  12. A Wireless Collaborated Inference Acceleration Framework for Plant Disease Recognition
  13. TWIST: Teleoperated Whole-Body Imitation System
  14. AOR: Anatomical Ontology-Guided Reasoning for Medical Large Multimodal Model in Chest X-Ray Interpretation
  15. Privacy Risks and Preservation Methods in Explainable Artificial Intelligence: A Scoping Review
  16. Towards Application-Specific Evaluation of Vision Models: Case Studies in Ecology and Biology
  17. MUSAR: Exploring Multi-Subject Customization from Single-Subject Dataset via Attention Routing
  18. ReplaceMe: Network Simplification via Depth Pruning and Transformer Block Linearization
  19. Database-Agnostic Gait Enrollment using SetTransformers
  20. Towards Quantifying the Hessian Structure of Neural Networks
  21. DPNet: Dynamic Pooling Network for Tiny Object Detection
  22. Plan-Driven Adaptive Bidding for First-Price Auctions with Budget Constraints under Nonstationarity
  23. HSplitLoRA: A Heterogeneous Split Parameter-Efficient Fine-Tuning Framework for Large Language Models
  24. Unsupervised training of keypoint-agnostic descriptors for flexible retinal image registration
  25. Advances in Automated Fetal Brain MRI Segmentation and Biometry: Insights from the FeTA 2024 Challenge
  26. Local Markov Equivalence for PC-style Local Causal Discovery and Identification of Controlled Direct Effects
  27. Beyond the Monitor: Mixed Reality Visualization and Multimodal AI for Enhanced Digital Pathology Workflow
  28. Unsupervised Deep Learning-based Keypoint Localization Estimating Descriptor Matching Performance
  29. Giving Simulated Cells a Voice: Evolving Prompt-to-Intervention Models for Cellular Control
  30. Bye-bye, Bluebook? Automating Legal Drudgery With AI-Augmented Rule Following
  31. Platelet enumeration in dense aggregates
  32. The use of Artificial Intelligence for Intervention and Assessment in Individuals with ASD
  33. Using Knowledge Graphs to harvest datasets for efficient CLIP model training
  34. Cooperative Variance Estimation and Bayesian Neural Networks for Disentangling Aleatoric and Epistemic Uncertainties
  35. Knowledge Graphs for Enhancing Large Language Models in Entity Disambiguation
  36. A Rate-Quality Model for Learned Video Coding
  37. Less is More: Efficient Weight Farcasting with 1-Layer Neural Network
  38. Voila: Voice-Language Foundation Models for Real-Time Autonomous Interaction and Voice Role-Play
  39. Multi-View Learning with Context-Guided Receptance for Image Denoising
  40. VGLD: Visually-Guided Linguistic Disambiguation for Monocular Depth Scale Recovery
  41. Structure Causal Models and LLMs Integration in Medical Visual Question Answering
  42. AI Standardized Patient Improves Human Conversations in Advanced Cancer Care
  43. Predicting Movie Hits Before They Happen with LLMs
  44. fastabx: A library for efficient computation of ABX discriminability
  45. Dance of Fireworks: An Interactive Broadcast Gymnastics Training System Based on Pose Estimation
  46. Multimodal Deep Learning for Stroke Prediction and Detection using Retinal Imaging and Clinical Data
  47. Grasp the Graph (GtG) 2.0: Ensemble of Graph Neural Networks for High-Precision Grasp Pose Detection in Clutter
  48. A Note on Statistically Accurate Tabular Data Generation Using Large Language Models
  49. Proper Noun Diacritization for Arabic Wikipedia: A Benchmark Dataset
  50. SCFormer: Structured Channel-wise Transformer with Cumulative Historical State for Multivariate Time Series Forecasting
  51. Sim2Real in endoscopy segmentation with a novel structure aware image translation
  52. Eye Movements as Indicators of Deception: A Machine Learning Approach
  53. Adaptive Budgeted Multi-Armed Bandits for IoT with Dynamic Resource Constraints
  54. Enhancing Chemical Reaction and Retrosynthesis Prediction with Large Language Model and Dual-task Learning
  55. Mallows-type model averaging: Non-asymptotic analysis and all-subset combination
  56. DeepSparse: A Foundation Model for Sparse-View CBCT Reconstruction
  57. A Theoretical Analysis of Provable Compositional Generalization in Neural Networks: A Necessary and Sufficient Condition
  58. LLaMA-Omni2: LLM-based Real-time Spoken Chatbot with Autoregressive Streaming Speech Synthesis
  59. Mirror Mean-Field Langevin Dynamics
  60. Automatic Proficiency Assessment in L2 English Learners
  61. Entropic Mirror Descent for Linear Systems: Polyak's Stepsize and Implicit Bias
  62. Lane-Wise Highway Anomaly Detection
  63. Study of the influence of a biased database on the prediction of standard algorithms for selecting the best candidate for an interview
  64. Connecting Independently Trained Modes via Layer-Wise Connectivity
  65. DELTA: Dense Depth from Events and LiDAR using Transformer's Attention
  66. Ensemble Kalman filter for uncertainty in human language comprehension
  67. RGBX-DiffusionDet: A Framework for Multi-Modal RGB-X Object Detection Using DiffusionDet
  68. Towards Cross-Modality Modeling for Time Series Analytics: A Survey in the LLM Era
  69. EMORL: Ensemble Multi-Objective Reinforcement Learning for Efficient and Flexible LLM Fine-Tuning
  70. Learning and Online Replication of Grasp Forces from Electromyography Signals for Prosthetic Finger Control
  71. Rethinking Federated Graph Learning: A Data Condensation Perspective
  72. Unified Multimodal Understanding and Generation Models: Advances, Challenges, and Opportunities
  73. Robustness questions the interpretability of graph neural networks: what to do?
  74. Bielik v3 Small: Technical Report
  75. Robust Duality Learning for Unsupervised Visible-Infrared Person Re-Identification
  76. Lazy But Effective: Collaborative Personalized Federated Learning with Heterogeneous Data
  77. Marker-Based Extrinsic Calibration Method for Accurate Multi-Camera 3D Reconstruction
  78. Advancing Constrained Monotonic Neural Networks: Achieving Universal Approximation Beyond Bounded Activations
  79. RobSurv: Vector Quantization-Based Multi-Modal Learning for Robust Cancer Survival Prediction
  80. Text to Image Generation and Editing: A Survey
  81. Bemba Speech Translation: Exploring a Low-Resource African Language
  82. Machine-Learning-Powered Neural Interfaces for Smart Prosthetics and Diagnostics
  83. FedSDAF: Leveraging Source Domain Awareness for Enhanced Federated Domain Generalization
  84. Uncovering Population PK Covariates from VAE-Generated Latent Spaces
  85. Exploring Design Choices for Autoregressive Deep Learning Climate Models
  86. Unveiling the Landscape of LLM Deployment in the Wild: An Empirical Study
  87. Corr2Distrib: Making Ambiguous Correspondences an Ally to Predict Reliable 6D Pose Distributions
  88. Bayesian Robust Aggregation for Federated Learning
  89. Integrating Column Generation and Large Neighborhood Search for Bus Driver Scheduling with Complex Break Constraints
  90. Automated Hybrid Reward Scheduling via Large Language Models for Robotic Skill Learning
  91. Finger Pose Estimation for Under-screen Fingerprint Sensor
  92. Point Cloud Recombination: Systematic Real Data Augmentation Using Robotic Targets for LiDAR Perception Validation
  93. Ming-Lite-Uni: Advancements in Unified Architecture for Natural Multimodal Interaction
  94. Deep learning of personalized priors from past MRI scans enables fast, quality-enhanced point-of-care MRI with low-cost systems
  95. Efficient Continual Learning in Keyword Spotting using Binary Neural Networks
  96. Timing Is Everything: Finding the Optimal Fusion Points in Multimodal Medical Imaging
  97. Data Augmentation With Back translation for Low Resource languages: A case of English and Luganda
  98. Incentivizing Inclusive Contributions in Model Sharing Markets
  99. Colombian Waitresses y Jueces canadienses: Gender and Country Biases in Occupation Recommendations from LLMs
  100. Recent Advances in Out-of-Distribution Detection with CLIP-Like Models: A Survey
  101. Investigating the Impact of Personalized AI Tutors on Language Learning Performance
  102. MSFNet-CPD: Multi-Scale Cross-Modal Fusion Network for Crop Pest Detection
  103. Towards Machine Learning-based Model Predictive Control for HVAC Control in Multi-Context Buildings at Scale via Ensemble Learning
  104. A New Approach to Backtracking Counterfactual Explanations: A Unified Causal Framework for Efficient Model Interpretability
  105. Towards One-shot Federated Learning: Advances, Challenges, and Future Directions
  106. Bielik 11B v2 Technical Report
  107. Token Coordinated Prompt Attention is Needed for Visual Prompting
  108. Estimating Commonsense Scene Composition on Belief Scene Graphs
  109. A probabilistic view on Riemannian machine learning models for SPD matrices
  110. Diagnostic Uncertainty in Pneumonia Detection using CNN MobileNetV2 and CNN from Scratch
  111. Uncertainty-Weighted Fusion of Image and Synthetic Event for Video Anomaly Detection
  112. Quantitative Analysis of Performance Drop in DeepSeek Model Quantization
  113. MetaScenes: Towards Automated Replica Creation for Real-world 3D Scans
  114. An Arbitrary-Modal Fusion Network for Volumetric Cranial Nerves Tract Segmentation
  115. Connecting Thompson Sampling and UCB: Towards More Efficient Trade-offs Between Privacy and Regret
  116. SuperEdit: Rectifying and Facilitating Supervision for Instruction-Based Image Editing
  117. Sharpness-Aware Minimization with Z-Score Gradient Filtering
  118. JTCSE: Joint Tensor-Modulus Constraints and Cross-Attention for Unsupervised Contrastive Learning of Sentence Embeddings
  119. Quaternion Sparse Decomposition for Multi-focus Color Image Fusion
  120. Quaternion Infrared Visible Image Fusion
  121. Advancing Email Spam Detection: Leveraging Zero-Shot Learning and Large Language Models
  122. Learning simple heuristic rules for classifying materials based on chemical composition
  123. Catastrophic Overfitting, Entropy Gap and Participation Ratio: A Noiseless lp Norm Solution for Fast Adversarial Training
  124. Social Biases in Knowledge Representations of Wikidata separates Global North from Global South
  125. Approximating Signed Distance Fields With Sparse Ellipsoidal Radial Basis Function Networks: A Dynamic Multi-Objective Optimization Strategy
  126. Temporal Robustness in Discrete Time Linear Dynamical Systems
  127. 6D Pose Estimation on Spoons and Hands
  128. VAEmo: Efficient Representation Learning for Visual-Audio Emotion with Knowledge Injection
  129. NeuroSim V1.5: Improved Software Backbone for Benchmarking Compute-in-Memory Accelerators with Device and Circuit-level Non-idealities
  130. What Is AI Safety? What Do We Want It to Be?
  131. Invoke Interfaces Only When Needed: Adaptive Invocation for Large Language Models in Question Answering
  132. Optimizing LLMs for Resource-Constrained Environments: A Survey of Model Compression Techniques
  133. Enabling Local Neural Operators to perform Equation-Free System-Level Analysis
  134. SafeMate: A Modular RAG-Based Agent for Context-Aware Emergency Guidance
  135. Generative Sign-description Prompts with Multi-positive Contrastive Learning for Sign Language Recognition
  136. ASAT: Adaptive Scoring and Thresholding with Human Feedback for Robust Out-of-Distribution Detection
  137. Entropy-Guided Sampling of Flat Modes in Discrete Spaces