AMC 2023

The 'AMC 2023' dataset/benchmark is used to evaluate the performance of reinforcement learning methods, specifically in the context of stabilizing large language model reasoning through techniques like Quantile Advantage Estimation.

Papers using AMC 2023 (2)

Stable Reinforcement Learning for Efficient Reasoning2025 · 25 citesQuantile Advantage Estimation: Stabilizing RLVR for LLM Reasoning2025