Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

OpenMath

Benchmarks

Task NameDataset NameSOTA ResultTrend
Mathematical ReasoningOpenMath-Instruct
Accuracy (Easy)81.022
4
Cross-file dependency reuse analysisOpenMath
Formalization-reflected Edges16
3
Semantic Correctness EvaluationOpenMath
Faithfulness54
2
Showing 3 of 3 rows