List your company

Scientific and Mathematical Reasoning on OlympiadBench (test)

99.63Math AccuracyHEART
77.592483.313789.03594.7563Sep 26, 2025

Evaluation results

MethodLinks
HEARTModel=Gemini 3 Pro, Prompt Strategy=HEART, Setting=Human-in-the-Loop Proxy2025.0999.6392.88
Self ReflectionModel=Gemini 3 Pro, Prompt Strategy=Self Reflection, Setting=Human-in-the-Loop Proxy2025.0999.4484.1
WaitModel=Gemini 3 Pro, Prompt Strategy=Wait, Setting=Human-in-the-Loop Proxy2025.0999.4488.28
CoTModel=Gemini 3 Pro, Prompt Strategy=CoT, Setting=Human-in-the-Loop Proxy2025.0999.2687.38
VanillaModel=Gemini 3 Pro, Prompt Strategy=Vanilla, Setting=Human-in-the-Loop Proxy2025.0994.3379.79
HEARTModel=Gemini 3 Flash, Prompt Strategy=HEART, Setting=Human-in-the-Loop Proxy2025.0992.4182.94
WaitModel=Gemini 3 Flash, Prompt Strategy=Wait, Setting=Human-in-the-Loop Proxy2025.0989.0775.13
Self ReflectionModel=Gemini 3 Flash, Prompt Strategy=Self Reflection, Setting=Human-in-the-Loop Proxy2025.0988.5271.96
CoTModel=Gemini 3 Flash, Prompt Strategy=CoT, Setting=Human-in-the-Loop Proxy2025.0987.4172.94
VanillaModel=Gemini 3 Flash, Prompt Strategy=Vanilla, Setting=Human-in-the-Loop Proxy2025.0978.4465.08