DemoEvolve: Overcoming Sparse Feedback in Agentic Harness Evolution with Demonstrations 文章

ArXiv CS.AI2026-05-26NEWSen作者: Lirong Che, Yuzhe yang, Peiwen lin, Chuang wang, Xueqian wang, Jian su

详细信息

来源站点
ArXiv CS.AI
作者
Lirong Che, Yuzhe yang, Peiwen lin, Chuang wang, Xueqian wang, Jian su
文章类型
NEWS
语言
en
发布日期
2026-05-26

摘要

arXiv:2605.24539v1 Announce Type: new Abstract: Agent harness evolution improves frozen language-model agents by modifying the executable structures around them. We study this paradigm as a form of sample-efficient fast adaptation: instead of updating model weights, an agent can acquire task-specific competence by changing its external harness, while leaving the base model's general capabilities intact. Prior work shows that self-generated rollouts can support harness search, suggesting that agents may acquire new task competence through practice. Yet in long-horizon stochastic environments, self-practice becomes fragile: rewards are sparse, outcomes are high-variance, and failures are hard to attribute to concrete harness mechanisms. We introduce DemoEvolve, a demonstration-bootstrapped approach to harness evolution.