AIME

The AIME dataset/benchmark is used to evaluate mathematical reasoning tasks.

Papers using AIME (4)

OpenThoughts: Data Recipes for Reasoning Models2025 · 1 citesSBSC: Step-By-Step Coding for Improving Mathematical Olympiad Performance2025 · 1 cites$V_1$: Unifying Generation and Self-Verification for Parallel Reasoners2026To Code or not to Code? Adaptive Tool Integration for Math Language Models via Expectation-Maximization2025