Search More, Think Less: Rethinking Long-Horizon Agentic Search for Efficiency and GeneralizationDate: 2026-02-27Fetched: 2026-02-28T01:46:47.900310+00:00AuthorsQianben Chen, Tianrui Qin, King Zhu, Qiexiang Wang, Chengjun Yu, Shu Xu, Jiaqi Wu, Jiayu Zhang, Xinpeng Liu, Xin Gui, Jingyi Cao, Piaohong Wang, Dingfeng Shi, He Zhu, Tiannan Wang, Yuqing Wang, Maojia Song, Tianyu Zheng, Ge Zhang, Jian Yang, Jiaheng Liu, Minghao LiuLinksHFarXivPDFGitHub14Abstract中文摘要English名为SMTL的深度学习框架通过以并行证据获取替代顺序推理，改进了高效长程智能体搜索，在多个研究基准上达到了最先进的性能，同时将推理步骤减少了70.7%。
