Low-Cost Hard-Label Adversarial Attack with Theoretical Foundations

Jun Liu, University of Macau; National Institute of Informatics; Leo Yu Zhang, Griffith University; Fengpeng Li, University of Macau; Isao Echizen, University of Tokyo; National Institute of Informatics; Jiantao Zhou, University of Macau

Hard-label black-box attacks, relying solely on top-1 predictions, represent one of the most challenging yet practically threat models. Despite recent progress, existing approaches face two key limitations: (1) they overlook the critical role of initialization, focusing primarily on optimization strategies; and (2) they rely heavily on empirical heuristics without theoretical guarantees. To bridge this gap, we establish a unified theoretical framework showing that existing sign-flipping hard-label attacks can be understood as approximating the true gradient sign. Guided by this {principled analysis}, we propose a novel attack framework featuring a zero-query initialization strategy and a Pattern-Driven Optimization (PDO) algorithm. We provide theoretical guarantees that our initialization yields higher cosine similarity to the true gradient sign than random baselines, and our PDO module achieves significantly lower query complexity than baseline search methods. Extensive experiments across CIFAR-10, ImageNet, and ObjectNet—covering standard and adversarially trained models, commercial APIs, and CLIP models—demonstrate that our method consistently outperforms SOTA hard-label attacks in both success rate and efficiency, particularly under low query budgets. Furthermore, our method demonstrates robust generalization across corrupted data (ImageNet-C), biomedical images (PathMNIST), and dense prediction tasks such as segmentation. Notably, it bypasses the stateful defense Blacklight, achieving a 0% detection rate.

USENIX Security '26 Open Access Sponsored by
King Abdullah University of Science and Technology (KAUST)

Open Access Media

USENIX is committed to Open Access to the research presented at our events. Papers and proceedings are freely available to everyone once the event begins. Any video, audio, and/or slides that are posted after the event are also free and open to everyone. Support USENIX and our commitment to Open Access.

BibTeX
@inproceedings {320733,
author = {Jun Liu and Leo Yu Zhang and Fengpeng Li and Isao Echizen and Jiantao Zhou},
title = {{Low-Cost} {Hard-Label} Adversarial Attack with Theoretical Foundations},
booktitle = {35th USENIX Security Symposium (USENIX Security 26)},
year = {2026},
isbn = {978-1-939133-58-8},
address = {Baltimore, MD},
pages = {4861--4880},
url = {https://www.usenix.org/conference/usenixsecurity26/presentation/liu-jun},
publisher = {USENIX Association},
month = aug
}

Presentation Video