AREX：邁向遞歸自我改進的深度研究智能體

摘要

深入研究需要智能体找出同时满足多重约束的答案。发现此类答案成本高昂，而验证候选答案通常可分解为易于处理的逐约束校验。这种发现与验证之间的不对称性表明，研究智能体不应仅延长搜索时间，更应通过验证中间结果，并利用部分验证状态引导后续优化，以递归方式改进当前答案。我们提出AREX——一类递归自改进的深度研究智能体。AREX在内层研究循环中收集证据并构建临时答案，在外层自改进循环中逐约束审计答案，识别未解决的论点，并启动针对性后续研究。为支撑长期递归自改进，AREX学习了一种自主上下文更新工具，将不断增长的交互历史压缩为紧凑的改进状态，保留已验证证据和未解决约束，且无需依赖外部模型。我们通过智能体中期训练和长视域强化学习，在经验证的合成任务及高质量轨迹上训练AREX。为缓解长视域学习中稀疏的最终奖励问题，我们强调在获取决定性证据或纠正错误研究方向的关键步骤。我们实例化了一个密集4B模型和一个122B-A10B混合专家模型。在BrowseComp、WideSearch、DeepSearchQA、人类最后的考试（HLE）及其他推理与工具使用基准测试中，AREX显著优于同规模基线，并与使用更多激活参数的模型保持竞争力。

English

Deep research requires agents to find answers that jointly satisfy multiple constraints. Discovering such answers is costly, whereas verifying a candidate can often be decomposed into tractable constraint-wise checks. This discovery--verification asymmetry suggests that a research agent should do more than simply search longer: it should recursively improve its current answer by verifying intermediate results and using the partially verified state to guide subsequent refinement. We introduce AREX, a family of Recursively Self-Improving (RSI) deep research agents. AREX alternates between an inner research loop that gathers evidence and constructs a provisional answer, and an outer self-improvement loop that audits the answer constraint-wise, identifies unresolved claims, and launches targeted follow-up research. To sustain RSI over long horizons, AREX learns an autonomous context-update tool that compresses growing interaction history into a compact improvement state preserving verified evidence and unresolved constraints, without relying on an external model. We train AREX on verified synthetic tasks and high-quality trajectories through agentic mid-training and long-horizon reinforcement learning. To mitigate sparse final rewards during long horizon learning, we emphasize key steps where decisive evidence is acquired or erroneous research directions are corrected. We instantiate a dense 4B model and a 122B-A10B Mixture-of-Experts model. Across BrowseComp, WideSearch, DeepSearchQA, Humanity's Last Exam (HLE), and other reasoning and tool-use benchmarks, AREX substantially outperforms comparable-scale baselines and remains competitive with models using substantially more activated parameters.