Daily Papers of 2025-08-15

1. We-Math 2.0: A Versatile MathBook System for Incentivizing Visual Mathematical Reasoning 141 upvotes, #1 of 2025-08-15
2. NextStep-1: Toward Autoregressive Image Generation with Continuous Tokens at Scale 134 upvotes, #2 of 2025-08-15
3. PRELUDE: A Benchmark Designed to Require Global Comprehension and Reasoning over Long Contexts 65 upvotes, #3 of 2025-08-15
4. ToonComposer: Streamlining Cartoon Production with Generative Post-Keyframing 49 upvotes, #4 of 2025-08-15
5. UI-Venus Technical Report: Building High-performance UI Agents with RFT 38 upvotes, #5 of 2025-08-15
6. A Survey on Diffusion Language Models 32 upvotes, #6 of 2025-08-15
7. STream3R: Scalable Sequential 3D Reconstruction with Causal Transformer 30 upvotes, #7 of 2025-08-15
8. Puppeteer: Rig and Animate Your 3D Models 30 upvotes, #7 of 2025-08-15
9. Pass@k Training for Adaptively Balancing Exploration and Exploitation of Large Reasoning Models 24 upvotes, #9 of 2025-08-15
10. HumanSense: From Multimodal Perception to Empathetic Context-Aware Responses through Reasoning MLLMs 11 upvotes, #10 of 2025-08-15
11. Processing and acquisition traces in visual encoders: What does CLIP know about your camera? 7 upvotes, #11 of 2025-08-15
12. From Black Box to Transparency: Enhancing Automated Interpreting Assessment with Explainable AI in College Classrooms 3 upvotes, #12 of 2025-08-15
13. When Explainability Meets Privacy: An Investigation at the Intersection of Post-hoc Explainability and Differential Privacy in the Context of Natural Language Processing 0 upvotes, #13 of 2025-08-15

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.