AgentSociety Challenge: Designing LLM Agents for User Modeling and Recommendation on Web Platforms
Yuwei Yan, Yu Shang, Qingbin Zeng, Keyu Zhao, Zhiheng Zheng, Xuefei Ning, Tianji Wu, Shengen Yan, Yu Wang, Fengli Xu, Yong Li · 2025
Read AgentSociety Challenge: Designing LLM Agents for User Modeling and Recommendation on Web Platforms by Yuwei Yan, Yu Shang, Qingbin Zeng, Keyu Zhao, Zhiheng Zheng, Xuefei Ning, Tianji Wu, Shengen Yan, Yu Wang, Fengli Xu, Yong Li (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2502-18754.txt.
Description
Recommender systems are starting to reason about users rather than just tally clicks, and a new contest tests that skill head-on. Yuwei Yan and co-authors report the AgentSociety Challenge, the first Web Conference competition for language-model agents in user modeling and recommendation. Teams worked with data from Yelp, Amazon, and Goodreads inside a simulated platform.
Two tracks split the problem: modeling the user, and making the recommendation. The contest drew 295 teams and more than 1,400 submissions over thirty-seven days. Winning agents improved scores by roughly a fifth during development and held strong gains in the finals. The 2025 report open-sources the arena for future agent builders.
Catalog record
| CREATOR | Yuwei Yan, Yu Shang, Qingbin Zeng, Keyu Zhao, Zhiheng Zheng, Xuefei Ning, Tianji Wu, Shengen Yan, Yu Wang, Fengli Xu, Yong Li |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 20 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2502-18754