Daily Papers of 2025-06-13

  1. ReasonMed: A 370K Multi-Agent Generated Dataset for Advancing Medical Reasoning 91 upvotes, #1 of 2025-06-13
  2. Magistral 58 upvotes, #2 of 2025-06-13
  3. SWE-Factory: Your Automated Factory for Issue Resolution Training Data and Evaluation Benchmarks 50 upvotes, #3 of 2025-06-13
  4. Text-Aware Image Restoration with Diffusion Models 37 upvotes, #4 of 2025-06-13
  5. AniMaker: Automated Multi-Agent Animated Storytelling with MCTS-Driven Clip Generation 36 upvotes, #5 of 2025-06-13
  6. VRBench: A Benchmark for Multi-Step Reasoning in Long Narrative Videos 31 upvotes, #6 of 2025-06-13
  7. Discrete Audio Tokens: More Than a Survey! 29 upvotes, #7 of 2025-06-13
  8. Comment on The Illusion of Thinking: Understanding the Strengths and Limitations of Reasoning Models via the Lens of Problem Complexity 27 upvotes, #8 of 2025-06-13
  9. Ming-Omni: A Unified Multimodal Model for Perception and Generation 26 upvotes, #9 of 2025-06-13
  10. PosterCraft: Rethinking High-Quality Aesthetic Poster Generation in a Unified Framework 26 upvotes, #9 of 2025-06-13
  11. Fine-Grained Perturbation Guidance via Attention Head Selection 26 upvotes, #9 of 2025-06-13
  12. Domain2Vec: Vectorizing Datasets to Find the Optimal Data Mixture without Training 23 upvotes, #12 of 2025-06-13
  13. Optimus-3: Towards Generalist Multimodal Minecraft Agents with Scalable Task Experts 21 upvotes, #13 of 2025-06-13
  14. Resa: Transparent Reasoning Models via SAEs 20 upvotes, #14 of 2025-06-13
  15. VideoDeepResearch: Long Video Understanding With Agentic Tool Using 19 upvotes, #15 of 2025-06-13
  16. Build the web for agents, not agents for the web 19 upvotes, #15 of 2025-06-13
  17. AutoMind: Adaptive Knowledgeable Agent for Automated Data Science 18 upvotes, #17 of 2025-06-13
  18. ChineseHarm-Bench: A Chinese Harmful Content Detection Benchmark 12 upvotes, #18 of 2025-06-13
  19. What Makes a Good Natural Language Prompt? 10 upvotes, #19 of 2025-06-13
  20. LaTtE-Flow: Layerwise Timestep-Expert Flow-based Transformer 10 upvotes, #19 of 2025-06-13
  21. Compound AI Systems Optimization: A Survey of Methods, Challenges, and Future Directions 10 upvotes, #19 of 2025-06-13
  22. CreatiPoster: Towards Editable and Controllable Multi-Layer Graphic Design Generation 10 upvotes, #19 of 2025-06-13
  23. Attention, Please! Revisiting Attentive Probing for Masked Image Modeling 8 upvotes, #23 of 2025-06-13
  24. Eliciting Fine-Tuned Transformer Capabilities via Inference-Time Techniques 7 upvotes, #24 of 2025-06-13
  25. UniPre3D: Unified Pre-training of 3D Point Cloud Models with Cross-Modal Gaussian Splatting 7 upvotes, #24 of 2025-06-13
  26. NoLoCo: No-all-reduce Low Communication Training Method for Large Models 7 upvotes, #24 of 2025-06-13
  27. VerIF: Verification Engineering for Reinforcement Learning in Instruction Following 6 upvotes, #27 of 2025-06-13
  28. DreamActor-H1: High-Fidelity Human-Product Demonstration Video Generation via Motion-designed Diffusion Transformers 6 upvotes, #27 of 2025-06-13
  29. EmbodiedGen: Towards a Generative 3D World Engine for Embodied Intelligence 6 upvotes, #27 of 2025-06-13
  30. Decomposing MLP Activations into Interpretable Features via Semi-Nonnegative Matrix Factorization 6 upvotes, #27 of 2025-06-13
  31. Token Perturbation Guidance for Diffusion Models 5 upvotes, #31 of 2025-06-13
  32. Draft-based Approximate Inference for LLMs 4 upvotes, #32 of 2025-06-13
  33. StreamSplat: Towards Online Dynamic 3D Reconstruction from Uncalibrated Video Streams 4 upvotes, #32 of 2025-06-13
  34. LLM Unlearning Should Be Form-Independent 3 upvotes, #34 of 2025-06-13
  35. TeleMath: A Benchmark for Large Language Models in Telecom Mathematical Problem Solving 3 upvotes, #34 of 2025-06-13
  36. MCA-Bench: A Multimodal Benchmark for Evaluating CAPTCHA Robustness Against VLM-based Attacks 2 upvotes, #36 of 2025-06-13
  37. LaMP-Cap: Personalized Figure Caption Generation With Multimodal Figure Profiles 2 upvotes, #36 of 2025-06-13
  38. Breaking Data Silos: Towards Open and Scalable Mobility Foundation Models via Generative Continual Learning 2 upvotes, #36 of 2025-06-13
  39. Discovering Hierarchical Latent Capabilities of Language Models via Causal Representation Learning 2 upvotes, #36 of 2025-06-13
  40. Beyond True or False: Retrieval-Augmented Hierarchical Analysis of Nuanced Claims 2 upvotes, #36 of 2025-06-13
  41. TaxoAdapt: Aligning LLM-Based Multidimensional Taxonomy Construction to Evolving Research Corpora 2 upvotes, #36 of 2025-06-13

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.