Value as Memory: Online Value-Guided Post-training for General Robot Policies
机器人终于有了「任务进度记忆」！连续干活4小时，成功率超95%
This research introduces VaM, a reinforcement learning framework that uses recent value predictions as task progress memory to enhance robot autonomy. VaM enables robots to continuously adapt to new states during deployment by integrating offline and online learning. The method shows significant improvements in task success rates, with over 95% success in a 4-hour continuous box assembly task. The paper also highlights the effectiveness of value caching in distinguishing visually similar states and triggering corrective actions.
The paper introduces VaM (Value as Memory), a method that uses recent value predictions as task progress memory to improve robot autonomy in long-duration tasks. It demonstrates significant improvements in task success rates and recovery from failures, particularly in tasks involving deformable objects and contact-rich operations.
Source: Machine Heart — Robotics on WeChat · Read original article ↗
English translation unavailable. Choose the original article to read it.
What the source reports
Publisher-reported claims, with original evidence. These results have not been independently verified by RoboSignal.
Reported numbers
Reported success rate
95%
Reported success rate
>95%
duration
4 hours
duration
4 hours
- paper: not_reported
- code: not_reported
- weights: not_reported
- dataset: not_reported
What remains unknown
Not established in the collected evidence: Environment, Control, Data origin.
Reported performance applies to the described task. It does not establish general autonomy or deployment readiness.
Source excerpts and review record
Automatically extracted; no manual editorial approval recorded.
机器人终于有了「任务进度记忆」！连续干活4小时，成功率超95%
Open source S1
在纸盒装配任务中，VaM 连续自主运行 4 小时，任务成功率达到 95% 以上，展示了长时间运行中的稳定表现。
Open source S9
Source：Machine Heart — Robotics on WeChat · mp.weixin.qq.com