Mitigating Tail Narrowing in LLM Self-Improvement via Socratic-Guided Sampling

Yiwen Ding, Zhiheng Xi, Wei He 0024, Lizhuoyuan Lizhuoyuan, Yitao Zhai, Shi Xiaowei, Xunliang Cai, Tao Gui, Qi Zhang 0001, Xuanjing Huang 0001. Mitigating Tail Narrowing in LLM Self-Improvement via Socratic-Guided Sampling. In Luis Chiruzzo, Alan Ritter, Lu Wang, editors, Proceedings of the 2025 Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics: Human Language Technologies, NAACL 2025 - Volume 1: Long Papers, Albuquerque, New Mexico, USA, April 29 - May 4, 2025. pages 10627-10646, Association for Computational Linguistics, 2025. [doi]

Abstract

Abstract is missing.