Daily Papers of 2026-06-15

  1. OmniDirector: General Multi-Shot Camera Cloning without Cross-Paired Data 106 upvotes, #1 of 2026-06-15
  2. APPO: Agentic Procedural Policy Optimization 77 upvotes, #2 of 2026-06-15
  3. Memory is Reconstructed, Not Retrieved: Graph Memory for LLM Agents 73 upvotes, #3 of 2026-06-15
  4. Measuring Epistemic Resilience of LLMs Under Misleading Medical Context 56 upvotes, #4 of 2026-06-15
  5. From Chatbot to Digital Colleague: The Paradigm Shift Toward Persistent Autonomous AI 56 upvotes, #4 of 2026-06-15
  6. HarnessX: A Composable, Adaptive, and Evolvable Agent Harness Foundry 46 upvotes, #6 of 2026-06-15
  7. Orchestra-o1: Omnimodal Agent Orchestration 45 upvotes, #7 of 2026-06-15
  8. Rethinking RAG in Long Videos: What to Retrieve and How to Use It? 36 upvotes, #8 of 2026-06-15
  9. From AGI to ASI 35 upvotes, #9 of 2026-06-15
  10. OmniVideo-100K: A Dataset for Audio-Visual Reasoning through Structured Scripts and Evidence Chains 31 upvotes, #10 of 2026-06-15
  11. Smaller Models are Natural Explorers for Policy-Level Diversity in GRPO 26 upvotes, #11 of 2026-06-15
  12. Skip a Layer or Loop It? Learning Program-of-Layers in LLMs 24 upvotes, #12 of 2026-06-15
  13. RedAct: Redacting Agent Capability Traces for Procedural Skill Protection 23 upvotes, #13 of 2026-06-15
  14. LLM Agents Can See Code Repositories 20 upvotes, #14 of 2026-06-15
  15. RepFusion: Leveraging Multimodal Priors for Denoising in Representation Space 18 upvotes, #15 of 2026-06-15
  16. Pythagoras-Prover: Advancing Efficient Formal Proving via Augmented Lean Formalisation 16 upvotes, #16 of 2026-06-15
  17. Hy-Embodied-0.5-VLA: From Vision-Language-Action Models to a Real-World Robot Learning Stack 15 upvotes, #17 of 2026-06-15
  18. World Tracing: Generative Pixel-Aligned Geometry Beyond the Visible 14 upvotes, #18 of 2026-06-15
  19. iMaC: Translating Actions into Motion and Contact Images for Embodied World Models 13 upvotes, #19 of 2026-06-15
  20. The Arbiter Agent: Continually Monitoring Multi-Agent Conversations to Detect Emergent Misalignment 13 upvotes, #19 of 2026-06-15
  21. MBench: A Comprehensive Benchmark on Memory Capability for Video World Models 11 upvotes, #21 of 2026-06-15
  22. RhymeFlow: Training-Free Acceleration for Video Generation with Asynchronous Denoising Flow Scheduling 11 upvotes, #21 of 2026-06-15
  23. No Hidden Prompts Needed! You Can Game AI Peer Review with Presentation-Only Revisions 10 upvotes, #23 of 2026-06-15
  24. μ_0: A Scalable 3D Interaction-Trace World Model 10 upvotes, #23 of 2026-06-15
  25. The Hidden Power of Scaling Factor in LoRA Optimization 9 upvotes, #25 of 2026-06-15
  26. Avatar V: Scaling Video-Reference Avatar Video Generation 9 upvotes, #25 of 2026-06-15
  27. When is Your LLM Steerable? 8 upvotes, #27 of 2026-06-15
  28. VISTA: View-Consistent Self-Verified Training for GUI Grounding 8 upvotes, #27 of 2026-06-15
  29. ClinHallu: A Benchmark for Diagnosing Stage-Wise Hallucinations in Medical MLLM Reasoning 8 upvotes, #27 of 2026-06-15
  30. StepPO: Step-Aligned Policy Optimization for Agentic Reinforcement Learning 6 upvotes, #30 of 2026-06-15
  31. An Enigma of Artificial Reason: Investigating the Production-Evaluation Gap in Large Reasoning Models 6 upvotes, #30 of 2026-06-15
  32. LoSoNA: A Benchmark for Local Social Norm Adaptation in Group Conversations 6 upvotes, #30 of 2026-06-15
  33. P3D-Bench: Benchmarking MLLMs for Parametric 3D Generation and Structural Reasoning 5 upvotes, #33 of 2026-06-15
  34. APT: Action Expert Pretraining Improves Instruction Generalization of Vision-Language-Action Policies 5 upvotes, #33 of 2026-06-15
  35. Benchmarking AI Agents for Addressing Scientific Challenges Across Scales 5 upvotes, #33 of 2026-06-15
  36. Dense Supervision, Sparse Updates: On the Sparsity and Geometry of On-Policy Distillation 5 upvotes, #33 of 2026-06-15
  37. AFFORDANCE20Q: Evaluating Affordance Reasoning from Physical Properties 5 upvotes, #33 of 2026-06-15
  38. AlloSpatial: Agentic Harness Framework for Spatial Reasoning in Foundation Models 4 upvotes, #38 of 2026-06-15
  39. Two-Fidelity Best-Action Identification for Stochastic Minimax Tree 3 upvotes, #39 of 2026-06-15
  40. Statistically Reliable LLM-Based Ranking Evaluation via Prediction-Powered Inference 3 upvotes, #39 of 2026-06-15
  41. AdaSR: Adaptive Streaming Reasoning with Hierarchical Relative Policy Optimization 3 upvotes, #39 of 2026-06-15
  42. Steady-Forcing: Balancing Spatial Persistence and Motion Continuity in Long-Horizon Nature Video Diffusion 3 upvotes, #39 of 2026-06-15
  43. FVSpec: Real-World Property-Based Tests as Lean Challenges 2 upvotes, #43 of 2026-06-15
  44. ActiveMimic: Egocentric Video Pretraining with Active Perception 2 upvotes, #43 of 2026-06-15
  45. Quickest Detection of Hallucination Onset: Delay Bounds and Learned CUSUM Statistics 2 upvotes, #43 of 2026-06-15
  46. Squeeze-Release: Iterative Pruning with Exact Structural Minimization 2 upvotes, #43 of 2026-06-15
  47. CARVE: Certified Affordable Repair of Vetoed Maneuvers via Envelopes for Interactive Driving 1 upvotes, #47 of 2026-06-15
  48. WaveDiT: Distribution-Aware Wavelet Flow Matching for Efficient 3D Brain MRI Synthesis 1 upvotes, #47 of 2026-06-15

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.