S-SPPO: Semantic-Calibrated Self-Play Preference Optimization

By Xi-Wen Chen, Wen-Hui Zhu, Jingjing Wang, Pei-Jie Qiu, Zhi-Peng Wang, Huayu Li, Zhengxiao He, Xuan-Zhao Dong, Prayag Tiwari, Ming-Kun Xu and 6 more. 2026

Subfields: Large Language Models, Alignment & RLHF