Your daily digest

Research papers, made simple to understand

The AI and machine learning papers people are reading most, each one rewritten in plain words: the problem, the idea, what was found and why it matters.

Older

Friday, 20 November 2026

    • AI Agents

    WeVisForensics jointly trains an investigator agent and evolving forensic inspectors for AI video detection

    WeVisForensics uses a main investigator agent to adaptively select forensic inspections from a self-evolving library of visual and audio sub-agents, trained via dual-timescale policy optimization to detect AI-generated videos with 95.2% accuracy and grounded evidence.

    Paper: WeVisForensics: Self-Evolving Agent for Omni AI-Generated Video Detection · Yipeng Du, Yuxi Li, Ying Zhang and 4 others

    49 upvotes on Hugging Face8 min readRead the article

    • Robotics & Physical AI
    • AI Agents

    Mind2Act World benchmark diagnoses reasoning, execution, and their composition in robot manipulation

    Mind2Act World provides 30 simulated robotic manipulation tasks to separately and jointly evaluate history-based reasoning, precise physical execution, and their closed-loop composition, revealing that current systems struggle most with composing reasoning and execution.

    Paper: Mind2Act World: A Diagnostic Benchmark for Embodied Mind, Physical Action, and Mind-Action Composition in Robotic Manipulation · Yuzhi Huang, Jie Wu, Fanding Huang and 40 others

    45 upvotes on Hugging Face6 min readRead the article

    • Multimodal & Vision

    WeMM-Video sets new open-source standards in multimodal video understanding

    WeMM-Video introduces Chinese-English bilingual models (2B to 35B parameters) that unify text, image, audio, and video processing with a hybrid-thinking approach, achieving state-of-the-art open-source performance on over 120 benchmarks including complex audio-visual reasoning.

    Paper: WeMM-Video: Pushing the Frontiers of Open-Source Omni-Modal Video Understanding · Changli Tang, Jie Yang, Tianyi Wang and 9 others

    39 upvotes on Hugging Face6 min readRead the article

Browse earlier papers

18 days of papers in November 2026

November 2026
123456789101112131415161718192021222324252627282930

Has papersYou are reading

  1. WeVisForensics jointly trains an investigator agent and evolving forensic inspectors for AI video detectionYou are reading this day
  2. Jeveration generates images and dense predictions in one fast forward passand 2 more papers
  3. Xing 4 advances large language models on Ascend chips with efficient architecture and trainingand 2 more papers
  4. Sage S2.1 Flash: A 320B-parameter multimodal model excelling in scientific reasoning and agentic tasksand 2 more papers
  5. Agentic robotics uses foundation models plus harnesses for versatile physical task controland 2 more papers
  6. Direct Preference Optimization fine-tunes language models from human preferences without reinforcement learningand 2 more papers
  7. Layer normalization stabilizes and speeds up training, especially in recurrent networksand 2 more papers
  8. SO-Arena benchmarks System One models' decision capabilities and reliability across domainsand 2 more papers
  9. Structured textbook-based supervision improves mathematical reasoning in large language modelsand 2 more papers
  10. Tree-Ising optimizes sparse bit-level edits to fix reasoning loops in language modelsand 2 more papers
  11. Unified tactile representation improves dexterous robot manipulation across sensor typesand 2 more papers
  12. Gaussian Spacing Normalization improves feed-forward 3D Gaussian Splatting generalizationand 2 more papers
  13. VFlash enables fast, accurate visual plan decoding by jointly predicting entire primitivesand 2 more papers
  14. TurnWise improves streaming video RL by assigning credit at each decision turnand 1 more paper
  15. Routing specialized AI models boosts efficiency and performance across tasksand 2 more papers
  16. Unified taxonomy and literature system organize agentic computer graphics researchand 2 more papers
  17. FrameForge adapts video models for diverse conditional image generation with input-adaptive temporal relationsand 2 more papers
  18. Diffusion models generate high-quality images by learning to reverse noise additionand 2 more papers