Daily Papers of 2025-06-27

1. Mind2Web 2: Evaluating Agentic Search with Agent-as-a-Judge 45 upvotes, #1 of 2025-06-27
2. FaSTA^*: Fast-Slow Toolpath Agent with Subroutine Mining for Efficient Multi-turn Image Editing 38 upvotes, #2 of 2025-06-27
3. MADrive: Memory-Augmented Driving Scene Modeling 35 upvotes, #3 of 2025-06-27
4. WorldVLA: Towards Autoregressive Action World Model 35 upvotes, #3 of 2025-06-27
5. Where to find Grokking in LLM Pretraining? Monitor Memorization-to-Generalization without Test 26 upvotes, #5 of 2025-06-27
6. Learning to Skip the Middle Layers of Transformers 16 upvotes, #6 of 2025-06-27
7. SAM4D: Segment Anything in Camera and LiDAR Streams 13 upvotes, #7 of 2025-06-27
8. PhysRig: Differentiable Physics-Based Skinning and Rigging Framework for Realistic Articulated Object Modeling 9 upvotes, #8 of 2025-06-27
9. Whole-Body Conditioned Egocentric Video Prediction 9 upvotes, #8 of 2025-06-27
10. Arch-Router: Aligning LLM Routing with Human Preferences 8 upvotes, #10 of 2025-06-27
11. FairyGen: Storied Cartoon Video from a Single Child-Drawn Character 7 upvotes, #11 of 2025-06-27
12. An Agentic System for Rare Disease Diagnosis with Traceable Reasoning 6 upvotes, #12 of 2025-06-27
13. Generative Blocks World: Moving Things Around in Pictures 5 upvotes, #13 of 2025-06-27
14. DiLoCoX: A Low-Communication Large-Scale Training Framework for Decentralized Cluster 4 upvotes, #14 of 2025-06-27
15. DuaShepherd: Integrating Stepwise Correctness and Potential Rewards for Mathematical Reasoning 3 upvotes, #15 of 2025-06-27
16. HeurAgenix: Leveraging LLMs for Solving Complex Combinatorial Optimization Challenges 2 upvotes, #16 of 2025-06-27
17. MuseControlLite: Multifunctional Music Generation with Lightweight Conditioners 2 upvotes, #16 of 2025-06-27

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.