作者位次 #2/5ICLR 2026Poster· Deep Learning / Algorithms
KBVQ-MoE: KLT-guided SVD with Bias-Corrected Vector Quantization for MoE Large Language Models
Zukang Xu · Zhixiong Zhao · Xing Hu · Zhixuan Chen · Dawei Yang
首次获取该候选人的完整档案及 StartLux 评估共 20 积分；每位用户单独解锁，之后查看免费。主动刷新 20 积分。
实时查找电子邮箱，费用不超过20积分；未找到邮箱不扣费。
Zukang Xu · Zhixiong Zhao · Xing Hu · Zhixuan Chen · Dawei Yang
Zhixiong Zhao · Zukang Xu · Dawei Yang
Large language models (LLMs) have driven major progress in NLP, yet their substantial memory and compute demands still hinder practical deployment. Binarization can compress weights to 1 bit, fundamentally lowering compute and bandwidth cost. However, existing methods cannot address activation he...