Daily Papers of 2026-06-19

  1. Moebius: 0.2B Lightweight Image Inpainting Framework with 10B-Level Performance 136 upvotes, #1 of 2026-06-19
  2. DragMesh-2: Physically Plausible Dexterous Hand-Object Interaction with Articulated Objects 72 upvotes, #2 of 2026-06-19
  3. Multi-LCB: Extending LiveCodeBench to Multiple Programming Languages 58 upvotes, #3 of 2026-06-19
  4. Playful Agentic Robot Learning 48 upvotes, #4 of 2026-06-19
  5. Beyond Static Leaderboards: Predictive Validity for the Evaluation of LLM Agents 41 upvotes, #5 of 2026-06-19
  6. S-Agent: Spatial Tool-Use Elicits Reasoning for Spatial Intelligence 39 upvotes, #6 of 2026-06-19
  7. DF3DV-1K: A Large-Scale Dataset and Benchmark for Distractor-Free Novel View Synthesis 32 upvotes, #7 of 2026-06-19
  8. FreeStyle: Free Control of Style-Content Dual-Reference Generation from Community LoRA Mining 28 upvotes, #8 of 2026-06-19
  9. FlowBender: Feedback-Aware Training for Self-Correcting Conditional Flows 21 upvotes, #9 of 2026-06-19
  10. JanusMesh: Fast and Zero-Shot 3D Visual Illusion Generation via Cross-Space Denoising 20 upvotes, #10 of 2026-06-19
  11. ImageWAM: Do World Action Models Really Need Video Generation, or Just Image Editing? 19 upvotes, #11 of 2026-06-19
  12. Context-Aware RL for Agentic and Multimodal LLMs 16 upvotes, #12 of 2026-06-19
  13. Current World Models Lack a Persistent State Core 15 upvotes, #13 of 2026-06-19
  14. ENPIRE: Agentic Robot Policy Self-Improvement in the Real World 14 upvotes, #14 of 2026-06-19
  15. HumanScale: Egocentric Human Video Can Outperform Real-Robot Data for Embodied Pretraining 13 upvotes, #15 of 2026-06-19
  16. Thinking with Visual Grounding 11 upvotes, #16 of 2026-06-19
  17. FAPO: Fully Autonomous Prompt Optimization of Multi-Step LLM Pipelines 10 upvotes, #17 of 2026-06-19
  18. Understanding the Behaviors of Environment-aware Information Retrieval 9 upvotes, #18 of 2026-06-19
  19. Holo-World: Unified Camera, Object and Weather Control for Video World Model 9 upvotes, #18 of 2026-06-19
  20. Rethinking Shrinkage Bias in LLM FP4 Pretraining: Geometric Origin, Systemic Impact, and UFP4 Recipe 9 upvotes, #18 of 2026-06-19
  21. The FID Lottery: Quantifying Hidden Randomness in Generative-Model Evaluation 9 upvotes, #18 of 2026-06-19
  22. Freeing the Law with LOCUS: A Local Ordinance Corpus for the United States 7 upvotes, #22 of 2026-06-19
  23. LooseControlVideo: Directorial Video Control using Spatial Blocking 6 upvotes, #23 of 2026-06-19
  24. LedgerAgent: Structured State for Policy-Adherent Tool-Calling Agents 6 upvotes, #23 of 2026-06-19
  25. Taylor-Calibrate: Principled Initialization for Hybrid Linear Attention Distillation 5 upvotes, #25 of 2026-06-19
  26. LegalHalluLens: Typed Hallucination Auditing and Calibrated Multi-Agent Debate for Trustworthy Legal AI 5 upvotes, #25 of 2026-06-19
  27. Adaptive Volumetric Mechanical Property Fields Invariant to Resolution 5 upvotes, #25 of 2026-06-19
  28. Selective Synergistic Learning for Video Object-Centric Learning 4 upvotes, #28 of 2026-06-19
  29. Configurable Clinical Information Extraction with Agentic RAG: What Works, What Breaks, and Why 4 upvotes, #28 of 2026-06-19
  30. Duration Aware Scheduling for ASR Serving Under Workload Drift 3 upvotes, #30 of 2026-06-19
  31. The Data Manifold under the Microscope 3 upvotes, #30 of 2026-06-19
  32. No Resource, No Benchmarks, No Problem? Evaluating and Improving LLMs for Code Generation in No-Resource Languages 3 upvotes, #30 of 2026-06-19
  33. JAMER: Project-Level Code Framework Dataset and Benchmark on Professional Game Engines 3 upvotes, #30 of 2026-06-19
  34. ReSyn: A Generalized Recursive Regular Expression Synthesis Framework 2 upvotes, #34 of 2026-06-19

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.