Expanded benchmark
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
Expanded benchmark Total
90Caught Count
2
Expanded benchmark Misapplied theorem
13Caught Count
2
Expanded benchmark Circular reasoning
9Caught Count
2
Expanded benchmark Arithmetic error
19Caught Rate
2
Expanded benchmark
20Caught Count
2
Expanded benchmark Hidden premise
29Caught Rate
2