1 paper
Yusheng Zhao, Bowei He, Xue Liu +1
Large language model agents are increasingly used for scientific discovery. However, existing benchmarks mix law inference with programming, score data fitting instead of predictio…