Continuous Vision-Language-Action Co-Learning with Semantic-Physical Alignment for Behavioral Cloning
具有语义-物理对齐的连续视觉-语言-动作共学用于行为克隆
专题命中 其他安全 :alignment(title,abstract);分类 cs.AI
AI总结 本文提出CCoL框架,通过连续共学视觉、语言和动作,解决行为克隆中的语义-物理不一致问题,提升动作执行的准确性和鲁棒性。
Comments Accepted at AAAI 2026, the Project website is available at https://qhemu.github.io/CCoL/