simplevla-rl-scaling-vla-training

【效率与协作】Apply reinforcement learning to Vision-Language-Action models for robotic control, achieving 99% LIBERO task success and discovering novel manipulation strategies (pushcut) without task-specific reward engineering. Scales efficiently via parallelized trajectory sampling and outcome-based rewards.

独立 GitHub 快照 · 自动分类 · 以原作者文档为准