Daily Papers of 2025-10-08

  1. Less is More: Recursive Reasoning with Tiny Networks 378 upvotes, #1 of 2025-10-08
  2. In-the-Flow Agentic System Optimization for Effective Planning and Tool Use 83 upvotes, #2 of 2025-10-08
  3. Fathom-DeepResearch: Unlocking Long Horizon Information Retrieval and Synthesis for SLMs 70 upvotes, #3 of 2025-10-08
  4. TaTToo: Tool-Grounded Thinking PRM for Test-Time Scaling in Tabular Reasoning 61 upvotes, #4 of 2025-10-08
  5. Fast-dLLM v2: Efficient Block-Diffusion LLM 47 upvotes, #5 of 2025-10-08
  6. CoDA: Coding LM via Diffusion Adaptation 39 upvotes, #6 of 2025-10-08
  7. Drax: Speech Recognition with Discrete Flow Matching 24 upvotes, #7 of 2025-10-08
  8. BIRD-INTERACT: Re-imagining Text-to-SQL Evaluation for Large Language Models via Lens of Dynamic Interactions 21 upvotes, #8 of 2025-10-08
  9. MixReasoning: Switching Modes to Think 21 upvotes, #8 of 2025-10-08
  10. Scaling Code-Assisted Chain-of-Thoughts and Instructions for Model Reasoning 19 upvotes, #10 of 2025-10-08
  11. ShapeGen4D: Towards High Quality 4D Shape Generation from Videos 15 upvotes, #11 of 2025-10-08
  12. CCD: Mitigating Hallucinations in Radiology MLLMs via Clinical Contrastive Decoding 14 upvotes, #12 of 2025-10-08
  13. Presenting a Paper is an Art: Self-Improvement Aesthetic Agents for Academic Presentations 13 upvotes, #13 of 2025-10-08
  14. ASPO: Asymmetric Importance Sampling Policy Optimization 13 upvotes, #13 of 2025-10-08
  15. OneFlow: Concurrent Mixed-Modal and Interleaved Generation with Edit Flows 12 upvotes, #15 of 2025-10-08
  16. Discrete Diffusion Models with MLLMs for Unified Medical Multimodal Generation 10 upvotes, #16 of 2025-10-08
  17. GRACE: Generative Representation Learning via Contrastive Policy Optimization 9 upvotes, #17 of 2025-10-08
  18. Scalable In-context Ranking with Generative Models 8 upvotes, #18 of 2025-10-08
  19. Mixing Mechanisms: How Language Models Retrieve Bound Entities In-Context 8 upvotes, #18 of 2025-10-08
  20. Human3R: Everyone Everywhere All at Once 8 upvotes, #18 of 2025-10-08
  21. Let it Calm: Exploratory Annealed Decoding for Verifiable Reinforcement Learning 7 upvotes, #21 of 2025-10-08
  22. TensorBLEU: Vectorized GPU-based BLEU Score Implementation for Per-Sentence In-Training Evaluation 7 upvotes, #21 of 2025-10-08
  23. HoloScene: Simulation-Ready Interactive 3D Worlds from a Single Video 7 upvotes, #21 of 2025-10-08
  24. LightCache: Memory-Efficient, Training-Free Acceleration for Video Generation 6 upvotes, #24 of 2025-10-08
  25. AInstein: Assessing the Feasibility of AI-Generated Approaches to Research Problems 6 upvotes, #24 of 2025-10-08
  26. Refusal Falls off a Cliff: How Safety Alignment Fails in Reasoning? 6 upvotes, #24 of 2025-10-08
  27. Equilibrium Matching: Generative Modeling with Implicit Energy-Based Models 5 upvotes, #27 of 2025-10-08
  28. Margin Adaptive DPO: Leveraging Reward Model for Granular Control in Preference Optimization 5 upvotes, #27 of 2025-10-08
  29. Scientific Algorithm Discovery by Augmenting AlphaEvolve with Deep Research 5 upvotes, #27 of 2025-10-08
  30. Demystifying deep search: a holistic evaluation with hint-free multi-hop questions and factorised metrics 4 upvotes, #30 of 2025-10-08
  31. CARE: Cognitive-reasoning Augmented Reinforcement for Emotional Support Conversation 3 upvotes, #31 of 2025-10-08
  32. VeriGuard: Enhancing LLM Agent Safety via Verified Code Generation 3 upvotes, #31 of 2025-10-08
  33. EgoNight: Towards Egocentric Vision Understanding at Night with a Challenging Benchmark 3 upvotes, #31 of 2025-10-08
  34. On Code-Induced Reasoning in LLMs 2 upvotes, #34 of 2025-10-08
  35. DRIFT: Learning from Abundant User Dissatisfaction in Real-World Preference Learning 2 upvotes, #34 of 2025-10-08
  36. No Tokens Wasted: Leveraging Long Context in Biomedical Vision-Language Models 2 upvotes, #34 of 2025-10-08
  37. ChartAgent: A Multimodal Agent for Visually Grounded Reasoning in Complex Chart Question Answering 2 upvotes, #34 of 2025-10-08
  38. Verifier-free Test-Time Sampling for Vision Language Action Models 2 upvotes, #34 of 2025-10-08
  39. Revisiting Modeling and Evaluation Approaches in Speech Emotion Recognition: Considering Subjectivity of Annotators and Ambiguity of Emotions 2 upvotes, #34 of 2025-10-08
  40. Adaptive Pruning for Increased Robustness and Reduced Computational Overhead in Gaussian Process Accelerated Saddle Point Searches 2 upvotes, #34 of 2025-10-08
  41. Distributional Semantics Tracing: A Framework for Explaining Hallucinations in Large Language Models 2 upvotes, #34 of 2025-10-08
  42. Deforming Videos to Masks: Flow Matching for Referring Video Segmentation 2 upvotes, #34 of 2025-10-08
  43. Training Dynamics Impact Post-Training Quantization Robustness 2 upvotes, #34 of 2025-10-08
  44. MEMTRACK: Evaluating Long-Term Memory and State Tracking in Multi-Platform Dynamic Agent Environments 1 upvotes, #44 of 2025-10-08
  45. A Contextual Quality Reward Model for Reliable and Efficient Best-of-N Sampling 1 upvotes, #44 of 2025-10-08
  46. Benchmark It Yourself (BIY): Preparing a Dataset and Benchmarking AI Models for Scatterplot-Related Tasks 1 upvotes, #44 of 2025-10-08
  47. BACHI: Boundary-Aware Symbolic Chord Recognition Through Masked Iterative Decoding on Pop and Classical Music 1 upvotes, #44 of 2025-10-08
  48. SDQM: Synthetic Data Quality Metric for Object Detection Dataset Evaluation 1 upvotes, #44 of 2025-10-08
  49. HalluGuard: Evidence-Grounded Small Reasoning Models to Mitigate Hallucinations in Retrieval-Augmented Generation 1 upvotes, #49 of 2025-10-08
  50. The Valley of Code Reasoning: Scaling Knowledge Distillation of Large Language Models 2 upvotes, #49 of 2025-10-08
  51. DYMO-Hair: Generalizable Volumetric Dynamics Modeling for Robot Hair Manipulation 3 upvotes, #49 of 2025-10-08

Data: hysts-bot-data/daily-papers-stats and the Daily Papers API. Open data: tardellirs/paper-pulse-data. Sister project: Model Pulse, the download history of every model on the Hub.