English

Soft Quantization: Model Compression Via Weight Coupling

Machine Learning 2026-01-30 v1 Disordered Systems and Neural Networks

Abstract

We show that introducing short-range attractive couplings between the weights of a neural network during training provides a novel avenue for model quantization. These couplings rapidly induce the discretization of a model's weight distribution, and they do so in a mixed-precision manner despite only relying on two additional hyperparameters. We demonstrate that, within an appropriate range of hyperparameters, our "soft quantization'' scheme outperforms histogram-equalized post-training quantization on ResNet-20/CIFAR-10. Soft quantization provides both a new pipeline for the flexible compression of machine learning models and a new tool for investigating the trade-off between compression and generalization in high-dimensional loss landscapes.

Keywords

Cite

@article{arxiv.2601.21219,
  title  = {Soft Quantization: Model Compression Via Weight Coupling},
  author = {Daniel T. Bernstein and Luca Di Carlo and David Schwab},
  journal= {arXiv preprint arXiv:2601.21219},
  year   = {2026}
}

Comments

7 pages, 6 figures