arXiv cs.AI· Yuval Kansal, Karen Ullrich, Polina Kirichenko, Ilia Kulikov, Niraj K. Jha, Mohammad Pezeshki, Mark Ibrahim·· 2 小时前AI 评分50
Self-Direction 让智能体从错误中持久学习，成功率提升 8.9 至 17.0 个百分点
Don't Forget What You Just Learned: Self-Directed Repair Turns Agent Mistakes Into Lasting Lessons
AI 导读
研究者提出 Self-Direction 机制，智能体失败后反思轨迹并将教训蒸馏为约占基础模型参数量 0.02% 的低秩 adapter，存入持久库并自动路由到相关任务。基于 Qwen 系列开源模型，在 GAIA、Claw-Eval 和 Terminal-Bench 2.1 上整体成功率提升 8.9 至 17.0 个百分点，减少重复犯错，少量 adapter 还能泛化到数十个新任务。
正文 · 原文
Abstract:Intelligent agents should learn from their mistakes so that experience leads to continuous improvement. Yet today's agents recover from a mistake only temporarily: in-context reflections are easily forgotten across sessions and compete for limited context, while full-parameter fine-tuning is expensive and can interfere with existing capabilities. We introduce Self-Direction, a mechanism for agents to learn from mistakes that combines the benefits of temporary in-context retries and policy updates. When an agent fails a task, it reflects on its own trajectory in order to extract a lesson and immediately distills it into a lightweight low-rank adapter. By keeping the original weights frozen, Self-Direction prevents adapter training from overwriting base-model parameters while training additional parameters equivalent to ~0.02% of the base model's parameter count for each adapter. The agent then saves these adapters in a persistent library of lessons, automatically routing them to solve related problems later. Experiments with agents based on open-weight models from the Qwen family across open-ended research (GAIA), tool-use workflows (Claw-Eval), and terminal tasks (Terminal-Bench 2.1) show that our method improves overall success rates by 8.9 to 17.0 percentage points and performs competitively with frontier systems. We find that Self-Direction reduces repeated mistakes across future encounters of a task, nearly doubling the reliability of an agent's corrections. Learned lessons also transfer: a handful of adapters generalize to dozens of new tasks without additional training. Ultimately, this self-directed repair transforms everyday mistakes into lasting lessons.
| Subjects: | Artificial Intelligence (cs.AI); Machine Learning (cs.LG) |
| Cite as: | arXiv:2610.31846 [cs.AI] |
| (or arXiv:2610.31846v1 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2610.31846 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Yuval Kansal [view email]
[v1]
Thu, 22 Oct 2026 19:29:38 UTC (1,686 KB)
来源：arXiv cs.AI · arxiv.org