ResearchBenchmarksReasoning on GPQAFollow36.33RACortexDebate26.9729.431.8334.26Jul 5, 2025Evaluation ResultsMethodMethodLinksRACortexDebateType=OursType=Ours2025.0736.33GDType=Part DebateType=Part Debate2025.0732.67NDType=Part DebateType=Part Debate2025.0732.33PRDType=Full DebateType=Full Debate2025.0732ChatEvalType=Full DebateType=Full Debate2025.0731.33RECONCILEType=Full DebateType=Full Debate2025.0731MLDType=Full DebateType=Full Debate2025.0728.33MaVType=No DebateType=No Debate2025.0727.33