Physical Intelligence 用世界模型评估机器人策略
Across VLA checkpoints of varying performance on our table bussing benchmark, world model evals reproduce the real-world policy ranking.
AI 导读
在不同性能的 VLA checkpoint 上，针对我们的 table bussing 基准，世界模型评估复现了真实世界的策略排名。 相关性图：预测成功率 vs 真实成功率
正文
当前显示原文，中文全文译文尚未就绪。
Across VLA checkpoints of varying performance on our table bussing benchmark, world model evals reproduce the real-world policy ranking.
correlation plot: predicted vs ground truth success rate