AI学会自己教自己搜索，不用老师也能变强
训练AI搜索工具通常需要人类精心标注的“标准答案”或更强大模型的示范，成本高且难以规模化。这篇论文反其道而行：他们构建了一个可验证的搜索环境（DeepSearch-World），让AI自己尝试、失败、反思、再尝试，把成功的经验作为自己的训练数据。关键创新在于环境本身是确定性的——搜索和阅读结果可复现，AI能准确判断自己是否走对路。经过多轮自我迭代，一个9B参数的模型在多个搜索基准上达到甚至超越依赖更强模型蒸馏的同类系统。它不是你明天能用上的，但它展示了一条让AI通过自我实践持续进化的路径，可能改变未来AI训练的方式。
📄 原文摘要(英文)
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.