List your company

Logical Reasoning on BigBench Hard Formal Fallacies

58.2AccuracyOPRO
53.72854.88956.0557.211May 18, 2026

Evaluation results

MethodLinks
OPROevaluations=30 prompt evaluations2026.0558.2
ReElicitevaluations=30 prompt evaluations2026.0558.1
APEevaluations=30 prompt evaluations2026.0557.6
PromptBreederevaluations=30 prompt evaluations2026.0554.6
TextGradevaluations=30 prompt evaluations2026.0553.9