Embodied AI TopConf · ICLR2026

Self-Improving Vision-Language-Action Models with Data Generation via Residual RL

ICLR2026 / Vision-Language-Action Models

VLA感知数据
新标签打开 AlphaXiv
Self-Improving Vision-Language-Action Models with Data Generation via Residual RL figure
AlphaXiv 中文论文页面（可滚动查看）