Translation in progress Some entries still include text in the original language.
Failure-First·Indexed by this site · Originally published
SC-VLA：通过稀疏世界想象实现在线动作修正
[Daily Paper] Self-Correcting VLA: Online Action Refinement via Sparse World Imagination
AI summary
SC-VLA 通过稀疏世界想象预测任务进度与轨迹趋势，再用基于 SAC 的残差策略在线修正执行误差。论文报告其在 ManiSkill3 仿真中平均成功率 86%、较 GR00T N1.5 提升 9%，ARX5 真机成功率 71%；相较 π0 执行步数减少 43%，PegInsertion 成功率提升 28%。
Read originalfailurefirst.org