L-GreCo: Layerwise-Adaptive Gradient Compression for Efficient and Accurate Deep Learning

Jul 22, 2026·
Jiaqi Li
Jiaqi Li
· 0 min read
Schema of L-GreCo integration
Abstract
Review of paper writing approach, summarizing research themes and innovations, sharing paper submission experiences.
Date
Jul 22, 2026 8:30 AM — 9:15 AM
Event
L-GreCo: Layerwise-Adaptive Gradient Compression for Efficient and Accurate Deep Learning
Location

Room 101, Xinda Building B

events
Group Meeting Research Sharing Gradient Compression Distributed Training
Sina Weibo
Jiaqi Li
Authors
Jiaqi Li (he/him)
M.S. Student
Received B.E. in Computer Science and Technology from CUIT in June 2023. Currently pursuing Master’s degree in Artificial Intelligence at CUIT since 2024.