Daily Papers of 2025-12-29

1. Mindscape-Aware Retrieval Augmented Generation for Improved Long Context Understanding 108 upvotes, #1 of 2025-12-29
2. InsertAnywhere: Bridging 4D Scene Geometry and Diffusion Models for Realistic Video Object Insertion 95 upvotes, #2 of 2025-12-29
3. MAI-UI Technical Report: Real-World Centric Foundation GUI Agents 26 upvotes, #3 of 2025-12-29
4. UniPercept: Towards Unified Perceptual-Level Image Understanding across Aesthetics, Quality, Structure, and Texture 24 upvotes, #4 of 2025-12-29
5. TimeBill: Time-Budgeted Inference for Large Language Models 22 upvotes, #5 of 2025-12-29
6. Masking Teacher and Reinforcing Student for Distilling Vision-Language Models 18 upvotes, #6 of 2025-12-29
7. ProEdit: Inversion-based Editing From Prompts Done Right 17 upvotes, #7 of 2025-12-29
8. See Less, See Right: Bi-directional Perceptual Shaping For Multimodal Reasoning 13 upvotes, #8 of 2025-12-29
9. InSight-o3: Empowering Multimodal Foundation Models with Generalized Visual Search 11 upvotes, #9 of 2025-12-29
10. Omni-Weather: Unified Multimodal Foundation Model for Weather Generation and Understanding 11 upvotes, #9 of 2025-12-29
11. SWE-RM: Execution-free Feedback For Software Engineering Agents 9 upvotes, #11 of 2025-12-29
12. SlideTailor: Personalized Presentation Slide Generation for Scientific Papers 8 upvotes, #12 of 2025-12-29
13. SVBench: Evaluation of Video Generation Models on Social Reasoning 7 upvotes, #13 of 2025-12-29
14. Rethinking Sample Polarity in Reinforcement Learning with Verifiable Rewards 2 upvotes, #14 of 2025-12-29
15. A 58-Addition, Rank-23 Scheme for General 3x3 Matrix Multiplication 2 upvotes, #14 of 2025-12-29

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.