BridgeV2W: Bridging Video Generation Models to Embodied World Models via Embodiment Masks figure
AlphaXiv 中文概览(可滚动查看)