List your company

Automated Fact Checking on Coverbench

85.8Macro F1CAAFC
82.36883.25984.1585.041May 12, 2026

Evaluation results

MethodLinks
CAAFCBackbone Model=GPT-OSS-120B2026.0585.885.9
CAAFCBackbone Model=LLAMA3.3-70B2026.0585.485.5
CAAFCBackbone Model=GEMMA3-27B2026.0582.582.9