VideoCoCo: Code-as-CoT for Physically-Consistent Video Generation via an Agentic Dual-Engine System Paper • 2607.27380 • Published 14 days ago • 71
Mage-VL: An Efficient Codec-Native Streaming Multimodal Foundation Model Paper • 2607.24904 • Published 16 days ago • 37
TimeLens2: Generalist Video Temporal Grounding with Multimodal LLMs Paper • 2607.17423 • Published 24 days ago • 167
SWE-Pruner Pro: The Coder LLM Already Knows What to Prune Paper • 2607.18213 • Published 23 days ago • 79
RESOURCE2SKILL: Distilling Executable Agent Skills from Human-Created Multimodal Resources Paper • 2606.29538 • Published 27 days ago • 143
SearchEyes: Towards Frontier Multimodal Deep Search Intelligence via Search World Simulation Paper • 2607.05943 • Published Jul 7 • 15