ViPo-MLLM: Visual-Pose Multimodal LLM for Gloss-Free Sign Language Translation figure
AlphaXiv 中文概览(可滚动查看)