Daily Papers of 2026-11-13

  1. SO-Arena: Benchmarking the Capability Spectrum and Reliability of System One Models 72 upvotes, #1 of 2026-11-13
  2. BirdDreams: Joint Perspective and BEV Video Generation for a Plausible, Reactive Driving World Model 56 upvotes, #2 of 2026-11-13
  3. G-STAR: Group-Scaled Token Capacity for Deeply Compressed Autoregressive Generation 46 upvotes, #3 of 2026-11-13
  4. Duo: A Reciprocal World Model for Full-Duplex Interaction 46 upvotes, #3 of 2026-11-13
  5. PhysLoop: Harnessing LLMs' Physics Awareness for Self-Improving Video Generation Models 42 upvotes, #5 of 2026-11-13
  6. Open Vision Banana: A Fully Open Recipe for Generative Visual Perception 40 upvotes, #6 of 2026-11-13
  7. Actor, Judge, or Teacher? Understanding Decision Models in Long-Horizon Agents 38 upvotes, #7 of 2026-11-13
  8. Ropedia Xperience 2: A Gold Standard for Scaling Physical AI Environments and Behaviors 35 upvotes, #8 of 2026-11-13
  9. On the Design of Hyper-Connections: Simplification, Stability, and Scalability 30 upvotes, #9 of 2026-11-13
  10. TTB: Test-time Baking for Efficient Rendering of Decoder-only View Synthesis Models 27 upvotes, #10 of 2026-11-13
  11. Learning the Task or Learning the Verifier? Rethinking Criteria Exposure in Embodied Self-Improvement 21 upvotes, #11 of 2026-11-13
  12. Unbundling Multimodal Vision Encoding 20 upvotes, #12 of 2026-11-13
  13. WorldDraft: If Code Builds the World, Can Video Models Just Render It? 18 upvotes, #13 of 2026-11-13
  14. Factored One-sided Adam-Moment for Practical and Scalable SOAP 14 upvotes, #14 of 2026-11-13
  15. UniSpark: Exploring Hybrid Attention for Multimodal Understanding and Generation 13 upvotes, #15 of 2026-11-13
  16. Who Remembers What? Multi-Granularity Memory Extraction for Multi-User, Multi-Agent Interaction 11 upvotes, #16 of 2026-11-13
  17. Experience-Handoff Reinforcement Learning for Recursive Self-Improvement 10 upvotes, #17 of 2026-11-13
  18. Agent-R^2: Scaling Agents for Realistic Productivity Work via Real-Source Data Synthesis 8 upvotes, #18 of 2026-11-13
  19. The Value of Thinking: Learning When to Reason in Hybrid Decision Models 8 upvotes, #18 of 2026-11-13
  20. Back to Pixels: Lifting Latent Video Diffusion into Pixel Space 7 upvotes, #20 of 2026-11-13
  21. DeepStress: Stress-Testing Deep Search Agents 6 upvotes, #21 of 2026-11-13
  22. JustRUA: A Controlled Study of GPT-6 Robot-Use Agents 6 upvotes, #21 of 2026-11-13
  23. Humanized but Still True? Auditing Facts and Detection in AI-Text Humanizers 5 upvotes, #23 of 2026-11-13
  24. HiVe-Mem: Interleaving Horizontal Memory Expansion and Vertical Evidence Composition for Multimodal Agents 5 upvotes, #23 of 2026-11-13
  25. From Large Language Models to Efficient System One Models via Adaptive Computation 5 upvotes, #23 of 2026-11-13
  26. Evidence-Guided Recovery for AI Agents with Ambiguous Tool Outcomes and Partial Execution 4 upvotes, #26 of 2026-11-13
  27. Do Multimodal Models Decide Consistently on Equivalent Optical Spectra? 4 upvotes, #26 of 2026-11-13
  28. TurtLES: A Large-Scale Benchmark for Turbulent 3D Neural PDE Surrogates 4 upvotes, #26 of 2026-11-13
  29. Consensus Capture in Reference-Based UAV Video Stabilization: A Benchmark on Dynamic Urban Scenes 4 upvotes, #26 of 2026-11-13
  30. The Narrative Gap: Can LLMs Help Us Navigate Diverse Narratives Across Languages? 3 upvotes, #30 of 2026-11-13
  31. A Cross-Scenario Survey and Analysis of Self-Evolving Multimodal Agents 3 upvotes, #30 of 2026-11-13
  32. EgoX: Egocentric Cross-Embodiment Manipulation with Embodiment Dreaming 3 upvotes, #30 of 2026-11-13
  33. Structured Confidence Perturbation via Fisher Geometry for Adversarial Training 2 upvotes, #33 of 2026-11-13
  34. A frontier machine intelligence for human neuroimaging 2 upvotes, #33 of 2026-11-13

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.