作者位次 #6/10ACL 2026Poster· Natural Language Processing
Rethinking Data Mixing from the Perspective of Large Language Models
Yuanjian Xu · Tianze Sun · Changwei Xu · XinLong Zhao · Jianing Hao · Ran Chen · Yang Liu · Ruijie Xu · Stephen Chen · Guang Zhang
摘要
Data mixing strategy is essential for large language model (LLM) training. Empirical evidence shows that inappropriate strategies can significantly reduce generalization. Although recent methods have improved empirical performance, several fundamental questions remain open: what constitutes a dom...