ResearchBenchmarksProof By Contradiction on D (test)Follow95AccuracyGPT92.9293.469494.54Oct 10, 2025Evaluation ResultsMethodMethodLinksAccuracyGPTExperiment Setting=OVE...Experiment Setting=OVE-all2025.1095T5Experiment Setting=OVE...Experiment Setting=OVE-all2025.1093