Daily Papers of 2025-06-03

  1. Beyond the 80/20 Rule: High-Entropy Minority Tokens Drive Effective Reinforcement Learning for LLM Reasoning 144 upvotes, #1 of 2025-06-03
  2. SmolVLA: A Vision-Language-Action Model for Affordable and Efficient Robotics 86 upvotes, #2 of 2025-06-03
  3. REASONING GYM: Reasoning Environments for Reinforcement Learning with Verifiable Rewards 61 upvotes, #3 of 2025-06-03
  4. Taming LLMs by Scaling Learning Rates with Gradient Grouping 36 upvotes, #4 of 2025-06-03
  5. Temporal In-Context Fine-Tuning for Versatile Control of Video Diffusion Models 35 upvotes, #5 of 2025-06-03
  6. SRPO: Enhancing Multimodal LLM Reasoning via Reflection-Aware Reinforcement Learning 32 upvotes, #6 of 2025-06-03
  7. LoHoVLA: A Unified Vision-Language-Action Model for Long-Horizon Embodied Tasks 30 upvotes, #7 of 2025-06-03
  8. ARIA: Training Language Agents with Intention-Driven Reward Aggregation 29 upvotes, #8 of 2025-06-03
  9. ShapeLLM-Omni: A Native Multimodal LLM for 3D Generation and Understanding 28 upvotes, #9 of 2025-06-03
  10. Jigsaw-R1: A Study of Rule-based Visual Reinforcement Learning with Jigsaw Puzzles 25 upvotes, #10 of 2025-06-03
  11. Learning Video Generation for Robotic Manipulation with Collaborative Trajectory Control 24 upvotes, #11 of 2025-06-03
  12. AReaL: A Large-Scale Asynchronous Reinforcement Learning System for Language Reasoning 22 upvotes, #12 of 2025-06-03
  13. EarthMind: Towards Multi-Granular and Multi-Sensor Earth Observation with Large Multimodal Models 21 upvotes, #13 of 2025-06-03
  14. Unified Scaling Laws for Compressed Representations 18 upvotes, #14 of 2025-06-03
  15. MiCRo: Mixture Modeling and Context-aware Routing for Personalized Preference Learning 15 upvotes, #15 of 2025-06-03
  16. Incentivizing Reasoning for Advanced Instruction-Following of Large Language Models 15 upvotes, #15 of 2025-06-03
  17. From Token to Action: State Machine Reasoning to Mitigate Overthinking in Information Retrieval 13 upvotes, #17 of 2025-06-03
  18. IVY-FAKE: A Unified Explainable Framework and Benchmark for Image and Video AIGC Detection 13 upvotes, #17 of 2025-06-03
  19. Cora: Correspondence-aware image editing using few step diffusion 11 upvotes, #19 of 2025-06-03
  20. Reasoning Like an Economist: Post-Training on Economic Problems Induces Strategic Generalization in LLMs 11 upvotes, #19 of 2025-06-03
  21. WebChoreArena: Evaluating Web Browsing Agents on Realistic Tedious Web Tasks 10 upvotes, #21 of 2025-06-03
  22. Normalized Attention Guidance: Universal Negative Guidance for Diffusion Model 9 upvotes, #22 of 2025-06-03
  23. Darwin Godel Machine: Open-Ended Evolution of Self-Improving Agents 9 upvotes, #22 of 2025-06-03
  24. VisualSphinx: Large-Scale Synthetic Vision Logic Puzzles for RL 9 upvotes, #22 of 2025-06-03
  25. DyePack: Provably Flagging Test Set Contamination in LLMs Using Backdoors 8 upvotes, #25 of 2025-06-03
  26. CodeV-R1: Reasoning-Enhanced Verilog Generation 8 upvotes, #25 of 2025-06-03
  27. Stress-testing Machine Generated Text Detection: Shifting Language Models Writing Style to Fool Detectors 8 upvotes, #25 of 2025-06-03
  28. Learning from Videos for 3D World: Enhancing MLLMs with 3D Vision Geometry Priors 8 upvotes, #25 of 2025-06-03
  29. OWSM v4: Improving Open Whisper-Style Speech Models via Data Scaling and Cleaning 8 upvotes, #25 of 2025-06-03
  30. zip2zip: Inference-Time Adaptive Vocabularies for Language Models via Token Compression 7 upvotes, #30 of 2025-06-03
  31. Esoteric Language Models 7 upvotes, #30 of 2025-06-03
  32. VAU-R1: Advancing Video Anomaly Understanding via Reinforcement Fine-Tuning 6 upvotes, #32 of 2025-06-03
  33. Cascading Adversarial Bias from Injection to Distillation in Language Models 6 upvotes, #32 of 2025-06-03
  34. WHEN TO ACT, WHEN TO WAIT: Modeling Structural Trajectories for Intent Triggerability in Task-Oriented Dialogue 6 upvotes, #32 of 2025-06-03
  35. Stepsize anything: A unified learning rate schedule for budgeted-iteration training 5 upvotes, #35 of 2025-06-03
  36. Pro3D-Editor : A Progressive-Views Perspective for Consistent and Precise 3D Editing 5 upvotes, #35 of 2025-06-03
  37. SATA-BENCH: Select All That Apply Benchmark for Multiple Choice Questions 5 upvotes, #35 of 2025-06-03
  38. LLM in the Loop: Creating the PARADEHATE Dataset for Hate Speech Detoxification 5 upvotes, #35 of 2025-06-03
  39. OmniResponse: Online Multimodal Conversational Response Generation in Dyadic Interactions 4 upvotes, #39 of 2025-06-03
  40. ComposeAnything: Composite Object Priors for Text-to-Image Generation 4 upvotes, #39 of 2025-06-03
  41. RARE: Retrieval-Aware Robustness Evaluation for Retrieval-Augmented Generation Systems 4 upvotes, #39 of 2025-06-03
  42. From Guidelines to Practice: A New Paradigm for Arabic Language Model Evaluation 4 upvotes, #39 of 2025-06-03
  43. Plan and Budget: Effective and Efficient Test-Time Scaling on Large Language Model Reasoning 3 upvotes, #43 of 2025-06-03
  44. Frankentext: Stitching random text fragments into long-form narratives 3 upvotes, #43 of 2025-06-03
  45. Think Again! The Effect of Test-Time Compute on Preferences, Opinions, and Beliefs of Large Language Models 3 upvotes, #43 of 2025-06-03
  46. MaskSearch: A Universal Pre-Training Framework to Enhance Agentic Search Capability 3 upvotes, #43 of 2025-06-03
  47. SenseFlow: Scaling Distribution Matching for Flow-based Text-to-Image Distillation 3 upvotes, #43 of 2025-06-03
  48. Pitfalls in Evaluating Language Model Forecasters 3 upvotes, #43 of 2025-06-03
  49. SealQA: Raising the Bar for Reasoning in Search-Augmented Language Models 3 upvotes, #43 of 2025-06-03
  50. How Programming Concepts and Neurons Are Shared in Code Language Models 3 upvotes, #43 of 2025-06-03
  51. MIKU-PAL: An Automated and Standardized Multi-Modal Method for Speech Paralinguistic and Affect Labeling 2 upvotes, #51 of 2025-06-03
  52. Pixels Versus Priors: Controlling Knowledge Priors in Vision-Language Models through Visual Counterfacts 2 upvotes, #51 of 2025-06-03
  53. R1-Code-Interpreter: Training LLMs to Reason with Code via Supervised and Reinforcement Learning 2 upvotes, #51 of 2025-06-03
  54. BinauralFlow: A Causal and Streamable Approach for High-Quality Binaural Speech Synthesis with Flow Matching Models 2 upvotes, #51 of 2025-06-03
  55. Neuro2Semantic: A Transfer Learning Framework for Semantic Reconstruction of Continuous Language from Human Intracranial EEG 2 upvotes, #51 of 2025-06-03
  56. MagiCodec: Simple Masked Gaussian-Injected Codec for High-Fidelity Reconstruction and Generation 2 upvotes, #51 of 2025-06-03
  57. Massively Multilingual Adaptation of Large Language Models Using Bilingual Translation Data 2 upvotes, #51 of 2025-06-03
  58. CityLens: Benchmarking Large Language-Vision Models for Urban Socioeconomic Sensing 2 upvotes, #51 of 2025-06-03
  59. LIFT the Veil for the Truth: Principal Weights Emerge after Rank Reduction for Reasoning-Focused Supervised Fine-Tuning 2 upvotes, #51 of 2025-06-03
  60. Aligning VLM Assistants with Personalized Situated Cognition 2 upvotes, #51 of 2025-06-03
  61. Shuffle PatchMix Augmentation with Confidence-Margin Weighted Pseudo-Labels for Enhanced Source-Free Domain Adaptation 1 upvotes, #61 of 2025-06-03
  62. Synthesis of discrete-continuous quantum circuits with multimodal diffusion models 0 upvotes, #62 of 2025-06-03

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.