Daily Papers of 2026-05-05

  1. MolmoAct2: Action Reasoning Models for Real-world Deployment 286 upvotes, #1 of 2026-05-05
  2. From Context to Skills: Can Language Models Learn from Context Skillfully? 152 upvotes, #2 of 2026-05-05
  3. Hallucinations Undermine Trust; Metacognition is a Way Forward 22 upvotes, #3 of 2026-05-05
  4. Persistent Visual Memory: Sustaining Perception for Deep Generation in LVLMs 21 upvotes, #4 of 2026-05-05
  5. Repetition over Diversity: High-Signal Data Filtering for Sample-Efficient German Language Modeling 19 upvotes, #5 of 2026-05-05
  6. AcademiClaw: When Students Set Challenges for AI Agents 16 upvotes, #6 of 2026-05-05
  7. WindowsWorld: A Process-Centric Benchmark of Autonomous GUI Agents in Professional Cross-Application Environments 15 upvotes, #7 of 2026-05-05
  8. OceanPile: A Large-Scale Multimodal Ocean Corpus for Foundation Models 15 upvotes, #7 of 2026-05-05
  9. ComboStoc: Combinatorial Stochasticity for Diffusion Generative Models 13 upvotes, #9 of 2026-05-05
  10. T^2PO: Uncertainty-Guided Exploration Control for Stable Multi-Turn Agentic Reinforcement Learning 10 upvotes, #10 of 2026-05-05
  11. Generative Modeling with Orbit-Space Particle Flow Matching 9 upvotes, #11 of 2026-05-05
  12. PhysicianBench: Evaluating LLM Agents in Real-World EHR Environments 9 upvotes, #11 of 2026-05-05
  13. Motion-Aware Caching for Efficient Autoregressive Video Generation 8 upvotes, #13 of 2026-05-05
  14. Perceptual Flow Network for Visually Grounded Reasoning 6 upvotes, #14 of 2026-05-05
  15. HiL-Bench (Human-in-Loop Benchmark): Do Agents Know When to Ask for Help? 5 upvotes, #15 of 2026-05-05
  16. Hierarchical Abstract Tree for Cross-Document Retrieval-Augmented Generation 5 upvotes, #15 of 2026-05-05
  17. Linear-Time Global Visual Modeling without Explicit Attention 5 upvotes, #15 of 2026-05-05
  18. Prior-Aligned Data Cleaning for Tabular Foundation Models 4 upvotes, #18 of 2026-05-05
  19. Assessing Pancreatic Ductal Adenocarcinoma Vascular Invasion: the PDACVI Benchmark 4 upvotes, #18 of 2026-05-05
  20. Agentic AI Systems Should Be Designed as Marginal Token Allocators 4 upvotes, #18 of 2026-05-05
  21. Counting as a minimal probe of language model reliability 4 upvotes, #18 of 2026-05-05
  22. Linking spatial biology and clinical histology via Haiku 3 upvotes, #22 of 2026-05-05
  23. Code World Model Preparedness Report 3 upvotes, #22 of 2026-05-05
  24. BlenderRAG: High-Fidelity 3D Object Generation via Retrieval-Augmented Code Synthesis 2 upvotes, #24 of 2026-05-05
  25. A Hybrid Approach for Closing the Sim2real Appearance Gap in Game Engine Synthetic Datasets 2 upvotes, #24 of 2026-05-05

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.