Daily Papers of 2026-05-06

  1. ARIS: Autonomous Research via Adversarial Multi-Agent Collaboration 110 upvotes, #1 of 2026-05-06
  2. OpenSeeker-v2: Pushing the Limits of Search Agents with Informative and High-Difficulty Trajectories 64 upvotes, #2 of 2026-05-06
  3. Beyond SFT-to-RL: Pre-alignment via Black-Box On-Policy Distillation for Multimodal RL 47 upvotes, #3 of 2026-05-06
  4. X2SAM: Any Segmentation in Images and Videos 25 upvotes, #4 of 2026-05-06
  5. Video Generation with Predictive Latents 24 upvotes, #5 of 2026-05-06
  6. HeavySkill: Heavy Thinking as the Inner Skill in Agentic Harness 22 upvotes, #6 of 2026-05-06
  7. SymptomAI: Towards a Conversational AI Agent for Everyday Symptom Assessment 11 upvotes, #7 of 2026-05-06
  8. StableI2I: Spotting Unintended Changes in Image-to-Image Transition 11 upvotes, #7 of 2026-05-06
  9. SVGS: Enhancing Gaussian Splatting Using Primitives with Spatially Varying Colors 9 upvotes, #9 of 2026-05-06
  10. StateSMix: Online Lossless Compression via Mamba State Space Models and Sparse N-gram Context Mixing 8 upvotes, #10 of 2026-05-06
  11. Generate, Filter, Control, Replay: A Comprehensive Survey of Rollout Strategies for LLM Reinforcement Learning 8 upvotes, #10 of 2026-05-06
  12. Reinforcement Learning for LLM-based Multi-Agent Systems through Orchestration Traces 7 upvotes, #12 of 2026-05-06
  13. PatRe: A Full-Stage Office Action and Rebuttal Generation Benchmark for Patent Examination 7 upvotes, #12 of 2026-05-06
  14. Workspace-Bench 1.0: Benchmarking AI Agents on Workspace Tasks with Large-Scale File Dependencies 7 upvotes, #12 of 2026-05-06
  15. ESARBench: A Benchmark for Agentic UAV Embodied Search and Rescue 6 upvotes, #15 of 2026-05-06
  16. SplAttN: Bridging 2D and 3D with Gaussian Soft Splatting and Attention for Point Cloud Completion 6 upvotes, #15 of 2026-05-06
  17. TCDA: Thread-Constrained Discourse-Aware Modeling for Conversational Sentiment Quadruple Analysis 6 upvotes, #15 of 2026-05-06
  18. Skills-Coach: A Self-Evolving Skill Optimizer via Training-Free GRPO 5 upvotes, #18 of 2026-05-06
  19. A Benchmark for Interactive World Models with a Unified Action Generation Framework 5 upvotes, #18 of 2026-05-06
  20. Healthcare AI GYM for Medical Agents 4 upvotes, #20 of 2026-05-06
  21. How Fast Should a Model Commit to Supervision? Training Reasoning Models on the Tsallis Loss Continuum 3 upvotes, #21 of 2026-05-06
  22. Chain of Evidence: Pixel-Level Visual Attribution for Iterative Retrieval-Augmented Generation 3 upvotes, #21 of 2026-05-06
  23. The TTS-STT Flywheel: Synthetic Entity-Dense Audio Closes the Indic ASR Gap Where Commercial and Open-Source Systems Fail 3 upvotes, #21 of 2026-05-06

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.