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.
@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}
}