MFI-ResNet: Efficient ResNet Architecture Optimization via MeanFlow Compression and Selective Incubation

TL;DR

This paper introduces MFI-ResNet, a novel ResNet optimization method that compresses and selectively expands network stages using MeanFlow modules, achieving higher efficiency and accuracy on CIFAR datasets.

Contribution

The paper proposes a new ResNet variant that integrates MeanFlow modules for compression and selective incubation, enhancing parameter efficiency and discriminative performance.

Findings

Reduces ResNet-50 parameters by over 45%.

Improves accuracy slightly on CIFAR-10 and CIFAR-100.

Demonstrates the effectiveness of flow-based feature transformation.

Abstract

ResNet has achieved tremendous success in computer vision through its residual connection mechanism. ResNet can be viewed as a discretized form of ordinary differential equations (ODEs). From this perspective, the multiple residual blocks within a single ResNet stage essentially perform multi-step discrete iterations of the feature transformation for that stage. The recently proposed flow matching model, MeanFlow, enables one-step generative modeling by learning the mean velocity field to transform distributions. Inspired by this, we propose MeanFlow-Incubated ResNet (MFI-ResNet), which employs a compression-expansion strategy to jointly improve parameter efficiency and discriminative performance. In the compression phase, we simplify the multi-layer structure within each ResNet stage to one or two MeanFlow modules to construct a lightweight meta model. In the expansion phase, we apply a selective incubation strategy to the first three stages, expanding them to match the residual block configuration of the baseline ResNet model, while keeping the last stage in MeanFlow form, and fine-tune the incubated model. Experimental results show that on CIFAR-10 and CIFAR-100 datasets, MFI-ResNet achieves remarkable parameter efficiency, reducing parameters by 46.28% and 45.59% compared to ResNet-50, while still improving accuracy by 0.23% and 0.17%, respectively. This demonstrates that generative flow-fields can effectively characterize the feature transformation process in ResNet, providing a new perspective for understanding the relationship between generative modeling and discriminative learning.