arXivJunlin Yang, Dylan Zhang, Xiangchen Song, Qirun Dai, Xiao Liu, Yuen Chen, Aniket Vashishtha, Jing Shi, Chenhao Tan, Hao PengMon, May 25, 2026, 9:57 AM PDT
score 16.5
Test environment reveals AI agents struggle to discover true cause-and-effect relationships
Original: CausaLab: A Scalable Environment for Interactive Causal Discovery Toward AI Scientists
Source: arxiv.org ↗
Writing ELI5 summary…