Daily Papers of 2026-07-13

1. Video Generation Models are General-Purpose Vision Learners 81 upvotes, #1 of 2026-07-13
2. Long-Horizon-Terminal-Bench: Testing the Limits of Agents on Long-Horizon Terminal Tasks with Dense Reward-Based Grading 74 upvotes, #2 of 2026-07-13
3. Scalable Visual Pretraining for Language Intelligence 55 upvotes, #3 of 2026-07-13
4. Trust Region Policy Distillation 34 upvotes, #4 of 2026-07-13
5. KronQ: LLM Quantization via Kronecker-Factored Hessian 32 upvotes, #5 of 2026-07-13
6. From RGB Generation to Dense Field Readout: Pixel-Space Dense Prediction with Text-to-Image Models 20 upvotes, #6 of 2026-07-13
7. Self-Guided Test-Time Training for Long-Context LLMs 19 upvotes, #7 of 2026-07-13
8. Towards Mechanistically Understanding Why Memorized Knowledge Fails to Generalize in Large Language Model Finetuning 18 upvotes, #8 of 2026-07-13
9. MedPMC: A Systematic Framework for Scaling High-Fidelity Medical Multimodal Data for Foundation Models 14 upvotes, #9 of 2026-07-13
10. PanoWorld: Real-World Panoramic Generation 13 upvotes, #10 of 2026-07-13
11. Flow-ERD: Agent-type Aware Flow Matching with Entropy-Regularized Distillation for Diverse Traffic Simulation 12 upvotes, #11 of 2026-07-13
12. VaseMuseum: Digital Intelligent Museum for Ancient Greek Pottery 10 upvotes, #12 of 2026-07-13
13. A Sovereign, Open-Source Foundation Model for German and English 10 upvotes, #12 of 2026-07-13
14. Phone Segmentation and Recognition through Phonological Activation Mapping 9 upvotes, #14 of 2026-07-13

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.