PDRVLA: A Vision-Language-Action Model with Physical Dynamics Representations Paper • 2610.37381 • Published 3 days ago • 151
SenseNova-U1.5: Towards Native Unified Visual Intelligence Paper • 2609.11929 • Published Sep 10 • 167
ArtifactsBench: Bridging the Visual-Interactive Gap in LLM Code Generation Evaluation Paper • 2507.04952 • Published Jul 7, 2025 • 12
DeepPresenter: Environment-Grounded Reflection for Agentic Presentation Generation Paper • 2602.22839 • Published Feb 26 • 4
DeepPresenter: Environment-Grounded Reflection for Agentic Presentation Generation Paper • 2602.22839 • Published Feb 26 • 4
DeepPresenter: Environment-Grounded Reflection for Agentic Presentation Generation Paper • 2602.22839 • Published Feb 26 • 4