Daily Papers of 2025-09-04

1. Open Data Synthesis For Deep Research 64 upvotes, #1 of 2025-09-04
2. Robix: A Unified Model for Robot Interaction, Reasoning and Planning 45 upvotes, #2 of 2025-09-04
3. Beyond Correctness: Harmonizing Process and Outcome Rewards through RL Training 20 upvotes, #3 of 2025-09-04
4. LMEnt: A Suite for Analyzing Knowledge in Language Models from Pretraining Data to Representations 19 upvotes, #4 of 2025-09-04
5. Mixture of Global and Local Experts with Diffusion Transformer for Controllable Face Generation 16 upvotes, #5 of 2025-09-04
6. Planning with Reasoning using Vision Language World Model 16 upvotes, #5 of 2025-09-04
7. MOSAIC: Multi-Subject Personalized Generation via Correspondence-Aware Alignment and Disentanglement 11 upvotes, #7 of 2025-09-04
8. Manipulation as in Simulation: Enabling Accurate Geometry Perception in Robots 8 upvotes, #8 of 2025-09-04
9. SATQuest: A Verifier for Logical Reasoning Evaluation and Reinforcement Fine-Tuning of LLMs 3 upvotes, #9 of 2025-09-04

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.