ResearchBenchmarksAutomated Fact Checking on CoverbenchFollow85.8Macro F1CAAFC82.36883.25984.1585.041May 12, 2026Evaluation ResultsMethodMethodLinksMacro F1AccuracyCAAFCBackbone Model=GPT-OSS...Backbone Model=GPT-OSS-120B2026.0585.885.9CAAFCBackbone Model=LLAMA3....Backbone Model=LLAMA3.3-70B2026.0585.485.5CAAFCBackbone Model=GEMMA3-27BBackbone Model=GEMMA3-27B2026.0582.582.9