List your company

Automated Fact Checking on AVeriTeC

77.6Macro F1 ScoreCAAFC
73.9674.90575.8576.795May 12, 2026

Evaluation results

MethodLinks
CAAFCBackbone Model=GEMMA3-27B2026.0577.689.4
CAAFCBackbone Model=LLAMA3.3-70B2026.0576.489.2
CAAFCBackbone Model=GPT-OSS-120B2026.0574.189.3