作者位次 #2/5ICML 2026Poster· Deep Learning / Large Language Models
S-Quant: Rethinking Weight Quantization with Seed-Based Generation
Mingzi Wang · Lancheng Zou · Shuo Yin · Zhuolun He · Bei Yu
效率与系统OpenReview
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。
Mingzi Wang · Lancheng Zou · Shuo Yin · Zhuolun He · Bei Yu
Zehua Pei · Hui-Ling Zhen · Lancheng Zou · Xianzhi Yu · Wulong Liu · Sinno Jialin Pan · Mingxuan Yuan · Bei Yu
Scaling large language models (LLMs) improves performance but significantly increases inference costs, with feed-forward networks (FFNs) consuming the majority of computational resources. While Mixture-of-Experts (MoE) architectures can reduce this cost through sparse activation, restructuring ex...