ResearchBenchmarksHallucination Detection on MiniCheckFollow0.838AUCStacking0.781840.796420.8110.82558May 8, 2026Evaluation ResultsMethodMethodLinksAUCStackingAnalyzer=Qwen2.5-7BAnalyzer=Qwen2.5-7B2026.050.838StackingAnalyzer=Gemma-2-9BAnalyzer=Gemma-2-9B2026.050.835StackingAnalyzer=LLaMA-3-8BAnalyzer=LLaMA-3-8B2026.050.834StackingAnalyzer=LLaMA-3-3BAnalyzer=LLaMA-3-3B2026.050.809StackingAnalyzer=Pythia-1.4BAnalyzer=Pythia-1.4B2026.050.803StackingAnalyzer=Gemma-2-2BAnalyzer=Gemma-2-2B2026.050.798StackingAnalyzer=Qwen2.5-0.5BAnalyzer=Qwen2.5-0.5B2026.050.784