REINFORCE++: A Simple and Efficient Approach for Aligning Large Language Models
REINFORCE++: A Simple and Efficient Approach for Aligning Large Language Models
Jian Hu
2025 · DOI: 10.48550/arXiv.2501.03262
arXiv.org · 155 Zitierungen
REINFORCE++: A Simple and Efficient Approach for Aligning Large Language Models
Jian Hu