List your company

Factual Claim Detection on PoliClaim Inconsistent samples S_inc (test)

63.6AgreementExperts
31.8840.11548.3556.585Feb 16, 2024

Evaluation results

MethodLinks
ExpertsSubset Reference=GPT-3.5 consistency (dagger)2024.0263.691.99
ExpertsSubset Reference=GPT-4 consistency (double dagger)2024.0262.990.79
AFaCTABackbone=GPT-42024.0241.874.64
AFaCTABackbone=GPT-3.52024.0233.165.8