作者位次 #4/5ICML 2026Poster· Deep Learning / Large Language Models
One LR Doesn’t Fit All: Heavy-Tail Guided Layerwise Learning Rates for LLMs
Di He · Songjun Tu · Keyu Wang · Lu Yin · Shiwei Liu
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。
Di He · Songjun Tu · Keyu Wang · Lu Yin · Shiwei Liu
Mingyu Cao · Alvaro Correia · Christos Louizos · Shiwei Liu · Lu Yin
Guinan Su · Yanwu Yang · Li Shen · Lu Yin · Shiwei Liu · Jonas Geiping
Hanqi Yan · Xiangxiang Cui · Lu Yin · Jindong Gu · Paul Pu Liang · Yulan He · Yifei Wang
The success of vision-language models is primarily attributed to effective cross-modal alignment between vision and language. However, modality gaps persist even in well-aligned models and may be necessary for human perception, as evidenced by modality-specific phenomena such as visual texture an...
Ziyang Chen · Zhenxuan Huang · Yile Wang · Weiqin Wang · Lu Yin · Hui Huang
Traditional sentence embedding methods employ token-level contrastive learning on non-generative pre-trained models. Recently, there have emerged embedding methods based on generative large language models (LLMs). These methods either rely on fixed prompt templates or involve modifications to the...