Iris：攀登搜索前沿的端到端搜索智能体方法

AI 摘要

本文提出了一种构建、训练和评估搜索智能体的端到端方法，并发布了Iris-mini和Iris-pro系统。该系统通过自动构造多跳任务、过滤训练数据、结合监督微调和强化学习进行迭代优化，在BrowseComp、DeepSearchQA等基准上取得了开源搜索智能体的最佳性能。研究还强调了上下文管理对搜索性能的重要影响，并提供了可测量的对比评估。

正文节选

Climbing to the Search Frontier numbers,square,comma,sortcompress 1 Introduction Search agents extend language models beyond closed-form reasoning by allowing them to interact with external tools and retrieve information from dynamic environments. A capable search agent must not only reason about the question, but also decide what to search, how to interpret retrieved evidence, when to continue exploring, and when the available evidence is sufficient to answer. This makes search a fundamentally

发布时间：2026-09-07 12:00

抓取时间：2026-09-07 12:04

来源机构：arXiv