MATH, GSM8K, CollegeMath, DeepMind-Mathematics, OlympiadBench-Math, TheoremQA

Benchmarks

Task NameDataset NameSOTA ResultTrendResults

MATH, GSM8K, CollegeMath, DeepMind-Mathematics, OlympiadBench-Math, TheoremQA Aggregated

49.4Average Accuracy

3