Can LLMs Predict the Future? A Brier Score Analysis of Prediction Markets
The RMF benchmark contains 3,000 resolved binary questions across nine domains, and results show why forecasting scores should be interpreted alongside simple probability baselines and question composition: under this protocol, newer versions or higher model tiers do not consistently produce more accurate probabilities...
Yuan-Bo Li, Ze-Kun Li, Xiao-Yan Cong
· 0 citations