Expanded benchmark

Benchmarks

Task Name	Dataset Name	SOTA Result
Adversarial detection	Expanded benchmark Total	Caught Count90	2
Adversarial detection	Expanded benchmark Misapplied theorem	Caught Count13	2
Adversarial detection	Expanded benchmark Circular reasoning	Caught Count9	2
Adversarial detection	Expanded benchmark Arithmetic error	Caught Rate19	2
Adversarial detection	Expanded benchmark	Caught Count20	2
Adversarial detection	Expanded benchmark Hidden premise	Caught Rate29	2

Showing 6 of 6 rows