arXiv cs.AI· Shengguang Wu, Suning Huang, Cherie Ho, Ria Doshi, Hao Zhu, Serena Yeung-Levy·· 11 小时前AI 评分58
RoboArchitect：用持久记忆自动设计机器人智能体外壳
RoboArchitect: Automating the Design of Robotic Agents with Persistent Memory
AI 导读
RoboArchitect 让编码模型观察冻结机器人策略的 rollout、诊断失败并迭代修改其智能体外壳，在多个基准和策略上把未见场景平均成功率从 9.7% 提升到 43.3%（4.5 倍），未见任务从 12.7% 提升到 40.2%（3.2 倍），全程不更新任何模型权重。
正文 · 原文
Abstract:Recent work shows that a frozen robot policy performs far better inside an agent harness, where control code decides when to query the policy, consult vision tools, or command the arm directly. Yet today these harnesses are built by hand and redesigned for each new robot or benchmark. We show that harness design can be automated, offering a strong alternative to finetuning on a novel setup. In RoboArchitect, a coding model observes a robot's rollouts, diagnoses failures, and iteratively revises the harness around the frozen policy. The harness evolves on a set of episodes for its optimization and is evaluated on held-out scenes and tasks. Without updating any model weights, RoboArchitect raises average success from 9.7% to 43.3% on unseen scenes (4.5x) and from 12.7% to 40.2% on unseen tasks (3.2x), across multiple benchmarks and policies. The gains are largest where the benchmark is far from the policy's training data: wrapping a general-purpose policy, the evolved harness yields a specialist that often outperforms even the finetuned version. Intuitive control strategies are automatically designed, such as lifting a jammed arm clear or reaching in horizontally when an obstacle blocks the way from above. The harness also evolves its own persistent memory, carrying learned experience, such as effective grasp depths, into later decisions. These results suggest that substantial headroom in robotic manipulation lies in the agent harness, and its design can be effectively automated and optimized without weight updates.
| Comments: | Project Website: this https URL |
| Subjects: | Robotics (cs.RO); Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV) |
| Cite as: | arXiv:2610.42141 [cs.RO] |
| (or arXiv:2610.42141v1 [cs.RO] for this version) | |
| https://doi.org/10.48550/arXiv.2610.42141 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Shengguang Wu [view email]
[v1]
Fri, 30 Oct 2026 17:58:48 UTC (13,747 KB)
来源：arXiv cs.AI · arxiv.org