ResearchBenchmarksDiscrimination between Good Faith and Problematic agents (Summarization) on XSum 18.5:1Follow6.69Cohen's dTVD-MI-0.55881.32313.2055.0869Aug 7, 2025Evaluation ResultsMethodMethodLinksCohen's dTVD-MI2025.086.69JudgeReference availability...Reference availability=ref2025.083.39GPPM2025.082.85MI (DoE)2025.081.89Baseline2025.080.29JudgeReference availability...Reference availability=no ref2025.08-0.28