DiG-Flow: Discrepancy-Guided Flow Matching for Robust VLA Models Paper • 2512.01715 • Published 8 days ago • 9
Being-H0: Vision-Language-Action Pretraining from Large-Scale Human Videos Paper • 2507.15597 • Published Jul 21 • 34
Unified Multimodal Understanding via Byte-Pair Visual Encoding Paper • 2506.23639 • Published Jun 30 • 4