ResearchBenchmarksMathematical Reasoning on AQuA (Mean/Std Accuracy)Follow93.42Mean AccuracyPSAO.LLM88.771289.978191.18592.3919May 14, 2026Evaluation ResultsMethodMethodLinksMean AccuracyStd Dev AccuracyPSAO.LLMModel=Gemini-2-FlashModel=Gemini-2-Flash2026.0593.421.32COPROModel=Gemini-2-FlashModel=Gemini-2-Flash2026.0592.761.12MIPROModel=Gemini-2-FlashModel=Gemini-2-Flash2026.0588.951.54