ResearchBenchmarksAutomated Fact Checking on AVeriTeCFollow77.6Macro F1 ScoreCAAFC73.9674.90575.8576.795May 12, 2026Evaluation ResultsMethodMethodLinksMacro F1 ScoreAccuracyCAAFCBackbone Model=GEMMA3-27BBackbone Model=GEMMA3-27B2026.0577.689.4CAAFCBackbone Model=LLAMA3....Backbone Model=LLAMA3.3-70B2026.0576.489.2CAAFCBackbone Model=GPT-OSS...Backbone Model=GPT-OSS-120B2026.0574.189.3