作者位次 #4/10ICML 2026Poster· Deep Learning / Foundation Models
DTop-p MoE: Sparsity-Controlled Dynamic Top-p MoE for Foundation Model Pre-training
Can Jin · Hongwu Peng · Mingcan Xiang · Qixin Zhang · Xiangchi Yuan · Amit Hasan · Ohi Dibua · Yifan Gong · Yan Kang · Dimitris Metaxas