arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2602.14234cs.AIcs.CL
收藏

REDSearcher: 一种可扩展且成本效益高的长周期搜索智能体框架

REDSearcher: A Scalable and Cost-Efficient Framework for Long-Horizon Search Agents

Zheng Chu, Xiao Wang, Jack Hong, Huiming Fan, Yuqi Huang, Yue Yang, Guohai Xu, Chenxiao Zhao, Cheng Xiang, Shengchao Hu, Dongdong Kuang, Ming Liu, Bing Qin, Xing Yu

AI总结:

REDSearcher提出一种统一框架,通过任务合成、中期训练和后期训练优化长周期搜索智能体,实现高效且低成本的搜索任务解决。

摘要:

大语言模型正从通用知识引擎向现实问题求解器过渡,但优化它们以进行深度搜索任务仍具挑战性。核心瓶颈在于高质量搜索轨迹和奖励信号的极端稀疏性,源于可扩展长周期任务构建的困难以及涉及外部工具调用的交互密集型 rollout 的高成本。为解决这些挑战,我们提出了REDSearcher,一种统一框架,集成了复杂任务合成、中期训练和后期训练,以实现可扩展搜索智能体的优化。具体而言,REDSearcher引入了以下改进:(1)我们将任务合成框架为双约束优化,其中任务难度由图拓扑和证据分散精确控制,允许生成复杂、高质量的任务。(2)我们引入了工具增强查询,以鼓励主动使用工具而不是被动回忆。(3)在中期训练中,我们显著增强了核心原子能力,包括知识、规划和功能调用,从而大幅降低了收集高质量轨迹用于下游训练的成本。(4)我们构建了一个本地模拟环境,使强化学习实验能够实现快速、低成本的算法迭代。在纯文本和多模态搜索智能体基准测试中,我们的方法实现了最先进的性能。为了促进未来长周期搜索智能体的研究,我们将发布10000条高质量复杂文本搜索轨迹、5000条多模态轨迹和1000条文本RL查询集,并与代码和模型检查点一起发布。

英文摘要:

Large language models are transitioning from generalpurpose knowledge engines to realworld problem solvers, yet optimizing them for deep search tasks remains challenging. The central bottleneck lies in the extreme sparsity of highquality search trajectories and reward signals, arising from the difficulty of scalable longhorizon task construction and the high cost of interactionheavy rollouts involving external tool calls. To address these challenges, we propose REDSearcher, a unified framework that codesigns complex task synthesis, midtraining, and posttraining for scalable searchagent optimization. Specifically, REDSearcher introduces the following improvements: (1) We frame task synthesis as a dualconstrained optimization, where task difficulty is precisely governed by graph topology and evidence dispersion, allowing scalable generation of complex, highquality tasks. (2) We introduce toolaugmented queries to encourage proactive tool use rather than passive recall.(3) During midtraining, we strengthen core atomic capabilities knowledge, planning, and function calling substantially reducing the cost of collecting highquality trajectories for downstream training. (4) We build a local simulated environment that enables rapid, lowcost algorithmic iteration for reinforcement learning experiments. Across both textonly and multimodal searchagent benchmarks, our approach achieves stateoftheart performance. To facilitate future research on longhorizon search agents, we will release 10K highquality complex text search trajectories, 5K multimodal trajectories and 1K text RL query set, and together with code and model checkpoints.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑