arXiv：cs.AI· Hyun Jung Lee, Jungtaek Kim, Jongwon Jeong, Tae-Eui Kam, Donghyun Kim, Yong Jae Lee·· 6 小时前AI 评分43
EMHO：基于经验轨迹的具身智能体框架自优化
EMHO: EMbodied Agent Harness Optimization via Experience Traces
AI 导读
研究者提出 EMHO（EMbodied Agent Harness Optimization），一种在具身模型冻结的前提下，通过分析执行轨迹和框架历史来迭代改进智能体框架的自进化方法，并配套推出 EMHO-Merge 以用 episode 级增益指导单一共享框架跨子任务的联合优化。
正文
Abstract:Improving embodied agents often focuses on optimizing the underlying model through training, while the surrounding agent harness that controls planning, context, and tool use is typically engineered. We ask whether this harness can instead improve itself directly from experience traces under sparse environmental feedback. We propose EMbodied Agent Harness Optimization (EMHO), a self-evolving framework that keeps the embodied model frozen and iteratively revises its harness by analyzing execution trajectories and prior harness history. EMHO optimizes beyond skills or recovery prompts, modifying how the agent monitors progress, uses vision tools, grounds observations, and responds to failures. To support multiple subtasks with a single harness, we introduce EMHO-Merge, which addresses trade-offs in jointly optimizing a single shared harness across subtasks by using episode-level gains and losses to guide evidence-supported refinement of when and how revised behaviors are applied. We evaluate EMHO on EmbodiedBench across navigation and manipulation tasks, and EMHO consistently improves task success for both Qwen 9B and 27B models. Qualitative analysis shows that EMHO goes beyond recovering from failures and unproductive actions to reshape how the embodied agent interprets and interacts with its environment.
| Subjects: | Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2610.08432 [cs.AI] |
| (or arXiv:2610.08432v1 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2610.08432 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Hyunjung Lee [view email]
[v1]
Tue, 6 Oct 2026 14:30:42 UTC (2,263 KB)
来源：arXiv：cs.AI · arxiv.org