DAPO-Math-17K

The 'DAPO-Math-17K' dataset/benchmark contains a collection of mathematical problems designed to evaluate the reasoning capabilities of Large Language Models (LLMs) in the context of Reinforcement Learning with Verifiable Rewards (RLVR).

Papers using DAPO-Math-17K (3)

Overconfident Errors Need Stronger Correction: Asymmetric Confidence Penalties for Reinforcement Learning2026MATH-Beyond: A Benchmark for RL to Expand Beyond the Base Model2025LearnAlign: Data Selection for LLM Reinforcement Learning with Improved Gradient Alignment2025