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

LogicNLI

Benchmarks

Task NameDataset NameSOTA ResultTrend
Natural Language InferenceLogicNLI
Accuracy41
26
NL-to-FOL Syntax CorrectnessLogicNLI (test)
Syntax Correctness Rate99
26
First-Order Logic ReasoningLogicNLI
Pass@176.6
18
Logical ReasoningLogicNLI
Accuracy63.4
11
Showing 4 of 4 rows