arXiv cs.CL· Jian Xie, Tianhe Lin, Zilu Wang, Yuting Ning, Yuekun Yao, Tianci Xue, Zhehao Zhang, Zhongyang Li, Kai Zhang, Yufan Wu, Shijie Chen, Boyu Gou, Mingzhe Han, Yifei Wang, Vint Lee, Xinpeng Wei, Xiangjun Wang, Yu Su, Huan Sun·· 17 小时前AI 评分65
QUEST：用全合成任务训练前沿深度研究智能体
QUEST: Training Frontier Deep Research Agents with Fully Synthetic Tasks
AI 导读
研究团队发布 QUEST 系列开放模型（2B 至 35B），定位为通用深度研究智能体，覆盖事实查找、引用溯源与报告综合等长程搜索任务。其训练配方结合 mid-training、监督微调和强化学习，核心是基于统一评分标准树的数据合成流程，无需人工标注即可生成带可验证奖励的训练数据，并内置上下文管理机制。
正文 · 原文
Authors:Jian Xie, Tianhe Lin, Zilu Wang, Yuting Ning, Yuekun Yao, Tianci Xue, Zhehao Zhang, Zhongyang Li, Kai Zhang, Yufan Wu, Shijie Chen, Boyu Gou, Mingzhe Han, Yifei Wang, Vint Lee, Xinpeng Wei, Xiangjun Wang, Yu Su, Huan Sun
Abstract:Deep research agents extend the role of search engines from retrieving keyword-matched pages to synthesizing knowledge, fundamentally changing how humans interact with information. However, frontier systems remain proprietary, while existing open agents often generalize poorly across different task types, leaving unclear how to train a broadly capable deep research agent. We release QUEST, a family of open models (ranging from 2B to 35B) that serve as general-purpose deep research agents designed to handle a wide range of long-horizon search tasks, with strong capabilities in fact seeking, citation grounding, and report synthesis. To build QUEST, we propose an effective training recipe combining mid-training, supervised fine-tuning, and reinforcement learning. Central to this recipe is a curated data synthesis pipeline based on unified rubric trees, which applies to different task types and enables synthesizing training data with verifiable rewards without human annotation. In addition, QUEST incorporates a built-in context management mechanism that enables effective long-horizon reasoning and knowledge synthesis. Using only 8K synthesized tasks, QUEST approaches or even surpasses frontier closed-source agents across eight deep research benchmarks spanning diverse task types, and achieves the best overall performance among recent open-weight agents. We released everything: models, data, and training scripts.
| Comments: | NeurIPS 2026 |
| Subjects: | Computation and Language (cs.CL) |
| Cite as: | arXiv:2605.24218 [cs.CL] |
| (or arXiv:2605.24218v2 [cs.CL] for this version) | |
| https://doi.org/10.48550/arXiv.2605.24218 arXiv-issued DOI via DataCite |
Submission history
From: Jian Xie [view email]
[v1]
Fri, 22 May 2026 20:59:20 UTC (24,751 KB)
[v2]
Sun, 25 Oct 2026 22:21:07 UTC (22,953 KB)
来源：arXiv cs.CL · arxiv.org