English(EN) GEPA-Curator: Curated Learning for Self Improving Agents

新的GEPA-Curator框架通过过滤任务失败来增强AI代理的学习

作者 PulseAugur 编辑部 · [1 个来源] · 2026-10-28 04:00

研究人员推出GEPA-Curator，一个旨在改进自改进AI代理学习过程的框架。该系统解决了在复杂环境中代理失败可能产生的嘈杂或不可靠的学习信号问题。GEPA-Curator过滤掉不太可能提供有用学习信号的任务，并选择性地检索成功的过去轨迹，为提示优化器提供额外上下文。使用Gemini 3.1 Pro和Claude Opus 5等模型在GoBrowse和AppWorld等基准上的评估表明，GEPA-Curator可与现有方法相媲美或超越现有方法，性能提升高达21.2%。 AI

影响该框架通过优化AI代理从经验中学习的效率和可靠性，有可能提高其在复杂任务中的表现。

排序理由该集群描述了一个新框架及其在arXiv上的研究论文中提出的评估。[lever_c_demoted from research: ic=1 ai=1.0]

在 arXiv cs.CL 阅读 →

AI 生成摘要 · Google Gemini · 来自 1 个来源。我们如何撰写摘要 →

本文如何被排名

Signal score: 18 / 100; Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket: Tool; 该集群描述了一个新框架及其在arXiv上的研究论文中提出的评估。[lever_c_demoted from research: ic=1 ai=1.0]
Source corroboration: Single-source cluster; Only one publisher covered this so far. Single-source stories can still rank when the publisher is high-authority, but they lack cross-source corroboration.
Topics: paper, model release; Editorial topic classification. Feeds into how the story surfaces on /topic/<slug> hub pages and into the per-entity coverage mix.
AI-industry relevance: High; Clearly on-topic for AI-industry coverage.
Story freshness: Breaking (< 6h); Fresh story with cross-source coverage still developing. Ranking may shift as more sources report.

完整方法见我们的编辑标准。

报道来源 [1]

arXiv cs.CL TIER_1 English(EN) · Zack Yu, Nicholas Lee, Minjae Lee, Lakshya Agarwal, Luca Manolache, Wonjun Kang, Matei Zaharia, Michael W. Mahoney, Kurt Keutzer, Sehoon Kim, Amir Gholami · 2026-10-28 04:00

GEPA-Curator：用于自改进代理的精选学习

arXiv:2610.37651v1 Announce Type: new Abstract: Large Language Models (LLMs) are increasingly used in multi-turn agentic workloads that require sustained interaction with complex environments. Existing reflective prompt optimizers such as GEPA learn from failed rollouts, but agen…

本文如何被排名

报道来源 [1]

GEPA-Curator：用于自改进代理的精选学习

相关实体

相关话题