Agents' Last Exam 论文提出面向长程经济价值任务的智能体基准 ALE
Agents' Last Exam
arXiv 论文 Agents' Last Exam（ALE）提出一个评估 AI 智能体在长程、有经济价值、结果可验证的真实任务上的基准，指出广泛基准成绩未转化为经济上有意义的部署，问题主要在评估环节。
Authors:Yiyou Sun, Xinyang Han, Weichen Zhang, Yuanbo Pang, Tianyu Wang, Yuhan Cao, Yixiao Huang, Xiang Liu, Zonglin Di, Chris Duroiu, Haoyun Zhang, Jeffrey Lin, Weishu Zhang, Tyler Zeng, Ying Yan, Bo Liu, Hanson Wen, Mingyang Xu, Xiaoyuan Liu, Zimeng Chen, Zhaowei Zhang, Weiyan Shi, Amanda Dsouza, Vincent Sunn Chen, Jingxu Xie, Dewen Lei, Hanxuan Zhang, Yinhe Wang, Xinran Zhang, Chao Zhou, Sheldon Huang, Patrick Bryant, Carl Boettiger, Yamini Rangan, Bradley Rothenberg, Kyle Steinfeld, Arvind Rao, Tapio Schneider, Georgios Yannakakis, Laure Zanna, Kaan Ozbay, Ida Sim, Tarek Zohdi, George Em Karniadakis, Jack Gallant, Teresa Head-Gordon, Chris Yuhao Liu, Rongwang Hu, Yafei Cheng, Aras Bacho, Zengyi Qin, Alan Yuille, Yixiong Chen, Hao Liu, Lin Zeng, Bo Chen, Zhendong Li, Litian Gong, Shashank Muralidhar Bharadwaj, Thorben Jansen, Haixiang Tang, Jianhong Tu, Jixian Liu, Maojia Song, Ruheng Wang, Shengcao Cao, Yanxin Zhang, Yingxuan Yang, Zongzheng Zhang, Honglin Bao, Hyunjun Kim, Peiran Xu, Shuo Lu, Zheng Zhang, Zhonghua Wang, Aditya Prakash, Bohan Lyu, Chang Ma, Haoxuan Chen, Hengduan Fan, Molei Tao, Roman Sitelew, Russell Alan Poldrack, Yang Liu, Yang Liu, Yichen Wu, Yuanze Lin, Zhenglu Li, Zibu Wei, Zijiao Chen, Abderrahim Bendahi, Alexander Schubert, Chase Broderick, Chengzhi Liu, Haoran Sun, Haoyang Li, Haoyue Hua, Jiayuan Ding, Jing Huang, Shangding Gu
et al. (279 additional authors not shown)
Abstract:Recent AI systems have achieved strong results on a wide range of benchmarks, yet these gains have not translated into economically meaningful deployment across many professional domains. We argue that this gap is largely an evaluation problem: widely used benchmarks lack sustained performance measurement on real and economically valuable workflows. This paper introduces Agents' Last Exam (ALE), a benchmark designed to evaluate AI agents on long horizon, economically valuable, real world tasks with verifiable outcomes. Developed in collaboration with 250+ industry experts, ALE covers non-physical industries defined with reference to O*NET / SOC 2018 (the U.S. federal occupational taxonomy). It is organized around a task taxonomy with 55 sub fields grouped into 13 industry clusters covering 1K+ tasks. Current results show that the hardest tier remains far from saturated: across mainstream harness and backbone configurations, the average full pass rate is below 1%. ALE is designed as a living benchmark: its task pool grows continuously as new workflows and industries are onboarded. More broadly, ALE is intended not merely as another leaderboard, but as an instrument for closing the gap between benchmark success and GDP relevant impact.
| Comments: | Updated author list and affiliations; scientific content unchanged. Project website: this https URL Code: this https URL |
| Subjects: | Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Machine Learning (cs.LG) |
| Cite as: | arXiv:2606.05405 [cs.AI] |
| (or arXiv:2606.05405v4 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2606.05405 arXiv-issued DOI via DataCite |
Submission history
From: Yiyou Sun [view email]
[v1]
Wed, 3 Jun 2026 20:20:46 UTC (22,645 KB)
[v2]
Thu, 11 Jun 2026 10:09:39 UTC (21,947 KB)
[v3]
Fri, 9 Oct 2026 08:13:57 UTC (21,947 KB)
[v4]
Sun, 25 Oct 2026 01:06:09 UTC (21,949 KB)
来源：arXiv cs.AI · arxiv.org