LegoNet: Memory Footprint Reduction Through Block Weight Clustering

1 st Joseph Bingham, 2 nd Noah Green, 3 rd Saman Zonouz · 2026

Read LegoNet: Memory Footprint Reduction Through Block Weight Clustering by 1 st Joseph Bingham, 2 nd Noah Green, 3 rd Saman Zonouz (2026) 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-2603-06606.txt.

Description

Bingham and colleagues propose LegoNet, a compression technique constructing blocks of weights across the whole model regardless of layer type and clustering those induced blocks. Clustering weights by blocks instead of individual values, they compress a ResNet-50 model trained on Cifar-10 and ImageNet with just 32 4x4 blocks. That shrinks memory footprint by over 64 times without removing weights, changing architecture, losing accuracy, retraining, or data. They also show an arrangement of 16 4x4 blocks giving 128 times compression with less than 3 percent accuracy loss, all with no need for retraining or fine-tuning.

Catalog record

CREATOR1 st Joseph Bingham, 2 nd Noah Green, 3 rd Saman Zonouz
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE26402
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2603.06606

Browse the full library · About the Tower of Babel