Rethinking VLM Representation for VLA Initialization 文章

ArXiv CS.CV2026-05-26NEWSen作者: Weifeng Lin, Siyuan Huang, Hao Li, Tingwei Chen, Ruichuan An, Xinyu Wei, Jianbo Liu, Hongsheng Li
查看原文 →
关系图谱

Rethinking VLM Representation for VLA Initialization · 相关技术

相关技术

vision-language-action modelsembodied VQAFull FinetuneVision-Language ModelsLoRAVQAVLMVLA