# MetaShare：面向开放式智能体的失败感知集体自我改进

> 原文链接：https://robotics.hellobike.com/papers/2026-10-15-2610-21887
> 频道：papers | 更新日期：2026-10-15 | 标签：arxiv, cs.RO
> 核心摘要：MetaShare通过记录成败修改和跨谱系检索，帮助开放式智能体避开失败路径，并在多类任务实验中验证效果。

# MetaShare：面向开放式智能体的失败感知集体自我改进

::: v-pre
> 原文链接：[MetaShare: Failure-Aware Collective Self-Improvement for Open-Ended Agents](https://arxiv.org/abs/2610.21887)（arXiv 2610.21887 · 发布日期：2026-10-15）

**MetaShare: Failure-Aware Collective Self-Improvement for Open-Ended Agents**

MetaShare通过记录成败修改和跨谱系检索，帮助开放式智能体避开失败路径，并在多类任务实验中验证效果。

## 核心要点

- 开放式自改进智能体很少复用失败尝试，搜索可能收敛到冗余策略。
- 框架记录成功与失败的智能体修改，并附带有符号性能变化。
- 检索高表现策略和高置信失败，结合审计器补足类别并降低冗余。
- 实验中位性能持平或超过 DGM-Hyperagents，并以更少迭代达到自身最佳。

**验证范围与局限**：验证范围为论文评审、编码、数学评分、机器人奖励设计和财务审计风险评估实验。

## 英文摘要 (Abstract)

Humans often learn fastest not by repeating success, but by remembering which failures not to repeat. Open-ended self-improving agents can share experience across lineages, as in GEA, or edit their own improvement mechanisms, as in DGM-Hyperagents, yet failed improvement attempts are rarely retained as reusable cross-branch knowledge, and search can collapse onto redundant strategies. We introduce MetaShare, a failure-aware framework that records successful and unsuccessful agent modifications together with signed performance deltas. Before each generation, MetaShare retrieves diverse high-performing strategies alongside high-confidence failures from other lineages, letting agents reuse effective mechanisms while avoiding known dead ends. A task-blind auditor periodically injects absent or underrepresented mechanism categories, while embedding-based de-ranking limits redundancy during retrieval. Multiple meta-agents propose modifications in parallel at heterogeneous temperatures behind a deterministic selector that gates applicability, scope, compilability, and size. Experiments across paper review, coding, math grading, robotics reward design, and financial audit-risk assessment show that MetaShare's median matches or exceeds DGM-Hyperagents' final performance and reaches its own best-performing agent in 1.7--2.9$\times$ fewer evolution iterations.

作者：Vishal Bhutani, Prasang Gupta, Sumanth Chundru, Kevin Paul
:::

## 原始参考来源 (Sources)

- [arXiv 2610.21887：MetaShare: Failure-Aware Collective Self-Improvement for Open-Ended Agents](https://arxiv.org/abs/2610.21887)