# Improving Adversarial Transferability via Model Alignment

Source: [Improving Adversarial Transferability via Model Alignment](<https://www.sota2.com/research/paper/68fa74211e209131cea5fd43>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2025-10-26T03:02:53.994Z

Authors: Avery Ma, Amir-massoud Farahmand, Yangchen Pan, Philip Torr, Jindong Gu

Published: 2023-11-30T12:15:49.000Z

## Abstract

Neural networks are susceptible to adversarial perturbations that are transferable across different models. In this paper, we introduce a novel model alignment technique aimed at improving a given source model's ability in generating transferable adversarial perturbations. During the alignment process, the parameters of the source model are fine-tuned to minimize an alignment loss. This loss measures the divergence in the predictions between the source model and another, independently trained model, referred to as the witness model. To understand the effect of model alignment, we conduct a geometric analysis of the resulting changes in the loss landscape. Extensive experiments on the ImageNet dataset, using a variety of model architectures, demonstrate that perturbations generated from aligned source models exhibit significantly higher transferability than those from the original source model.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2311.18495>)

[PDF](<https://arxiv.org/pdf/2311.18495.pdf>)