Iris：攀登搜索前沿的代理模型

AI 摘要

Hugging Face 论文页面介绍了 Iris-mini 和 Iris-pro 两个搜索代理，分别基于 35B-A3B 和 397B-A17B 规模训练。它们通过结合监督微调和强化学习的多阶段流程，在 BrowseComp、DeepSearchQA 等基准上取得了开源搜索代理中的领先成绩。论文计划发布模型权重及完整训练配方。

正文节选

Iris: Climbing to the Search Frontier Abstract Two large-scale search agents are trained via a multi-stage pipeline combining supervised fine-tuning and reinforcement learning against live search, achieving state-of-the-art open-source results on complex web benchmarks through rigorous trajectory filtering and inference-time context management. We present Iris-mini and Iris-pro, two search agents trained at the 35B-A3B and 397B-A17B scales, together with the data pipeline and training recipe beh

发布时间：—

抓取时间：2026-09-07 10:29

来源机构：Hugging Face