DeepSearch-World：可验证环境中深度搜索智能体的自蒸馏

摘要

训练工具使用型智能体从自身经验中改进仍具挑战性，因为监督微调依赖于固定的教师蒸馏轨迹，而稀疏奖励强化学习在长程交互中提供弱监督信号。我们提出DeepSearch-Evolve——一个面向网络智能体的自蒸馏框架，其基础是DeepSearch-World：一个具备确定性验证能力、支持可复现搜索与页面阅读工具的环境。DeepSearch-World包含42万条基于实体级随机游走构建的多跳问答任务，支持自我进化所需的关键认知行为，包括进度验证、基于事实的反思及失败恢复。DeepSearch-Evolve通过迭代执行轨迹生成、筛选、数据混合与微调，训练出更强的智能体。无需从更强大模型进行蒸馏，DeepSearch-World-9B在开源智能体中展现出竞争性表现：BrowseComp达31.2%，GAIA达61.5%，HotpotQA达93.4%，表明可验证环境能够支撑长程网络智能体的可扩展自我进化。我们将开源环境、42万训练池、验证集、模型及代码，以促进未来关于自我改进型深度搜索智能体的研究。

English

Training tool-use agents to improve from their own experience remains challenging, as supervised fine-tuning relies on fixed teacher-distilled trajectories, while sparse-reward reinforcement learning provides weak supervision for long-horizon interactions. We present DeepSearch-Evolve, a self-distillation framework for web agents built on DeepSearch-World, a deterministic and verifiable environment with reproducible search and page-reading tools. DeepSearch-World contains 420K multi-hop QA tasks constructed from entity-level random walks and supports key agentic cognitive behaviors useful for self-evolving, including progress verification, grounded reflection, and failure recovery. DeepSearch-Evolve iteratively performs trajectory generation, filtering, data mixing, and fine-tuning to train stronger agents. Without distillation from more capable models, DeepSearch-World-9B achieves competitive performance compared with open-source agents, reaching 31.2% on BrowseComp, 61.5% on GAIA, and 93.4% on HotpotQA, showing that verifiable environments enable scalable self-evolution for long-horizon web agents. We will release the environment, 420K training pool, validation set, model, and code to facilitate future research on self-improving deep search agents.