一作ICML 2026Poster· General Machine Learning / Representation Learning
RQ-MoE: Residual Quantization via Mixture of Experts for Efficient Input-Dependent Vector Compression
Zhengjia Zhong · Shuyan Ke · Zaizhou Lin · Jiaqi Song · Hongyi Lan · Hui Li
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。
Zhengjia Zhong · Shuyan Ke · Zaizhou Lin · Jiaqi Song · Hongyi Lan · Hui Li
Hongyi Lan · Jiaqi Song · Zhengjia Zhong · Hui Li · Hong Liu · Xianming Lin · Rongrong Ji
Large Language Models (LLMs) demonstrate strong generation and reasoning abilities, but they still face challenges in long-term memory retention and multi-turn conversational consistency. Existing memory-augmented methods often incorporate full dialog histories without filtering, resulting in inf...