ReTrace: Cross-Stage Failure Attribution for Self-Improving Robotic Manipulation
ReTrace: Cross-Stage Failure Attribution for Self-Improving Robotic Manipulation
This paper presents ReTrace, a self-improvement framework for robotic manipulation that enables autonomous physical interaction and retrospective reasoning without provided demonstrations or model-parameter updates. The method traces downstream failures to earlier decisions, proposes targeted corrections, and evaluates them through subsequent interactions. Evaluated on four real-world tasks using a Franka Research 3 robot, ReTrace achieves significant improvements in success rates for plug insertion and ball pouring tasks.
Source: arXiv Robotics — research abstracts · Read original article ↗
Loading article text…
What the source reports
Publisher-reported claims, with original evidence. These results have not been independently verified by RoboSignal.
- Control
- Reported autonomousOpen source S2
Source excerpts and review record
Automatically extracted; no manual editorial approval recorded.
ross attempts, ReTrace progressively consolidates validated corrections into a reusable manipulation skill that guide subsequent execution. We evaluate ReTrace on four real-world manipulation tasks using a Franka Research 3 robot. On the two tasks with strong cross-stage dependencies, GPT-6 Astra-low improves from 0% to 80% success on plug insertion and from 0% to 100% on ball pouring, while GPT-6 Sol-xhigh improves
Open source S4
arXiv:2610.15814v1 Announce Type: new Abstract: General-purpose multimodal foundation models are increasingly capable of directly controlling robots, yet improving their manipulation capabilities through autonomous interaction remains challenging. A key challenge to such self-improvement is learning from failures in complex manipulation tasks involving cross-stage dependencies, where an action may succeed locally but
Open source S2
Source：arXiv Robotics — research abstracts · arxiv.org