# Yet Another Intermediate-Level Attack

Source: [Yet Another Intermediate-Level Attack](<https://www.sota2.com/research/paper/68fa68371e209131cea1ad74>)

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

Data updated: 2025-10-25T22:32:00.581Z

Authors: Qizhang Li, Yiwen Guo, Hao Chen

Published: 2020-08-20T09:14:04.000Z

## Abstract

The transferability of adversarial examples across deep neural network (DNN) models is the crux of a spectrum of black-box attacks. In this paper, we propose a novel method to enhance the black-box transferability of baseline adversarial examples. By establishing a linear mapping of the intermediate-level discrepancies (between a set of adversarial inputs and their benign counterparts) for predicting the evoked adversarial loss, we aim to take full advantage of the optimization procedure of multi-step baseline attacks. We conducted extensive experiments to verify the effectiveness of our method on CIFAR-100 and ImageNet. Experimental results demonstrate that it outperforms previous state-of-the-arts considerably. Our code is at https://github.com/qizhangli/ila-plus-plus.

## 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/2008.08847>)

[PDF](<https://arxiv.org/pdf/2008.08847.pdf>)