作者位次 #2/8ICML 2026Poster· Deep Learning / Large Language Models
AdaGC: Enhancing LLM Pretraining Stability via Adaptive Gradient Clipping
Guoxia Wang · Shuai Li · Congliang Chen · Jinle Zeng · Jiabin Yang · Dianhai Yu · Yanjun Ma · Li Shen
数据与预训练OpenReview