CURing Large Models: Compression via CUR Decomposition

Sanghyeon Park, Soo-Mook Moon · 2025

Read CURing Large Models: Compression via CUR Decomposition by Sanghyeon Park, Soo-Mook Moon (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2501-04211.txt.

Description

Sanghyeon Park and Soo-Mook Moon shrink large AI models in a 2025 paper. Big models perform well but eat memory and power. Their method, called CURing, rebuilds each weight grid from a few of its own columns and rows plus a small linking grid.

Rows and columns are picked by both their size and how active they are in use, so the most telling parts survive. On a leading open model with eight billion settings, the method cut the count by about nine percent in just over two minutes. That speed is more than twenty times faster than past methods, with little loss in skill.

Catalog record

CREATORSanghyeon Park, Soo-Mook Moon
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE75 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2501-04211

Browse the full library · About the Tower of Babel