DRA-GRPO: Your GRPO Needs to Know Diverse Reasoning Paths for Mathematical Reasoning
By Xi-Wen Chen, Wen-Hui Zhu, Pei-Jie Qiu, Xuan-Zhao Dong, Hao Wang, Hai-Yu Wu, Huayu Li, Aristeidis Sotiras, Yalin Wang, Abolfazl Razi. 2025
Subfields: Prompting & Reasoning, Alignment & RLHF, Reinforcement Learning