direct mujoco rendererGPT-5.5
Codex- Held-out 分数
- -11.81
- Checkpoint
- #007
- 重跑
- 50 steps
Validation-selected checkpoint rerun in the original environment; case return -12.00.
Robotics · Core16
抓取、抬起并把物体放到目标位置。
这些证据属于论文时期实验,并不表示相关 package 当前仍然可用。
每张卡片都在原研究 Environment 中重跑 validation 选出的 checkpoint。
direct mujoco rendererValidation-selected checkpoint rerun in the original environment; case return -12.00.
direct mujoco rendererValidation-selected checkpoint rerun in the original environment; case return -13.00.
direct mujoco rendererValidation-selected checkpoint rerun in the original environment; case return -13.00.
direct mujoco rendererValidation-selected checkpoint rerun in the original environment; case return -35.00.
在该 Environment 内分数越高越好;不同任务的原始 reward scale 不可比较。
-11.81-13.22-13.34-20.56-43.75