Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Log inSign up
Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Blog
Job Board
Q3AS

© 2026 Aqora Quantum S.A.S.

TermsPrivacyLegal Notice

Quantum-Aware Image Encoding and Adversarial Perturbation

Naziba Tasnim, Sk Saif Ibna Ezhar Arko, Adnan Hossain Sadi, Lail Subhe, Omar Nasif Chowdhury

Published Jan 1, 2026• 0 citations
Publisher
DOI
10.13140/rg.2.2.28665.63848

Related papers

Quantum Image Representations

Fei Yan, Salvador Elías Venegas-Andraca

Springer eBooks•Jan 1, 2020•5 citations

Quantum-Aware Image Encoding and Adversarial Perturbation

Naziba Tasnim, Sk Saif Ibna Ezhar Arko, Adnan Hossain Sadi +2 more

The emergence of generative AI poses problems for people in several industries, including artists and other content creators whose work can be trained on without their consent, along with public figures who can be impersonated. As generative AI is now being researched at the quantum level, adversarial AI in quantum remains a largely unexplored topic. With present NISQ architecture and current hardware limitations for quantum image processing in mind, we present a hybrid quantum adversarial AI system using the QPIXL++ library for investigating image perturbations with quantum encoding using Flexible Representation of Quantum Images (FRQI) and Sparse Fast Walsh Hadamard Transform (SFWHT) compression. Topics for investigation in this area include pixel-domain vs frequency-domain perturbations under quantum compression, optimal perturbation strength, quantum superposition, working around quantum hardware limitations and applying optimization techniques with models as DINOv2. This project aims to establish a foundation and launch new research directions in hybrid quantum machine learning and adversarial AI. After completion of the study, it was found that the mean preserved perturbation value for the images was 99.99%, meaning that most of the classically introduced perturbations survived quantum encoding. Additionally, the mean SSIM vaule was close to 1 and the mean PSNR value was 32.30 dB which is above the accepted threshold for indistinguishable images to human eyes.

•Jan 1, 2026•0 citations

Q-MALDetect: Quantum Image-Based Malware Detection and Adversarial Robustness Analysis

Walid El Maouaki, Mohammad Izen, Nouhaila Innan +2 more

Quantum machine-learning (QML) models are increasingly considered for security applications, yet their adversarial robustness in malware analysis remains insufficiently understood. This paper presents Q-MALDetect, a hybrid convolutional and variational-quantum detector for byte-image representations of x86-64 Executable and Linkable Format (ELF) binaries, and evaluates it under feature-space and a range of Grad-CAM-guided, executable-aware evasion attacks. The model achieves 97.2% training accuracy, but its robustness deteriorates as attack strength increases, with the strongest evaluated feature-space attack reaching a 78.7% attack success rate (ASR). Among executable-aware attacks, Grad-CAM-guided no-operation (NOP) insertion produces the highest conditional ASR of 32.9%. Dynamic adversarial training is the most effective evaluated defense, reducing feature-space ASR to as low as 5.0%, executable-aware ASR to 7.7%, and the sequential-attack ASR to 34.0%. In contrast, convolutional-autoencoder preprocessing increases executable-aware ASR to 39.7%, coinciding with a rise in Grad-CAM attribution mass within the mutable .text region from 35.2% to 45.7%. Ablation results reveal no monotonic relationship between qubit count and robustness, while the encoding interface modifies the learned attack surface without neutralizing the strongest attack. These findings demonstrate that the effectiveness of attacks and defenses depends strongly on the perturbation constraints and evaluation conditions. Reliable deployment of QML malware detectors therefore requires assessment against both feature-space and executable-aware attacks rather than reliance on clean accuracy or generic denoising alone.

cs.CRCross-listedarXiv:2610.26240•Oct 20, 2026

Controlled Steering-Based State Preparation for Adversarial-Robust Quantum Machine Learning

Sahan Sanjaya, Hari Krishna Parvatham, Emma Andrews +1 more

Quantum machine learning (QML) provides a promising framework for leveraging quantum-mechanical effects in learning tasks. However, its vulnerability to adversarial perturbations remains a major challenge for practical deployment. In QML systems, small perturbations applied to classical inputs can propagate through the quantum encoding stage and distort the resulting quantum state, thereby degrading model performance. In this work, we propose a defense mechanism that replaces the conventional quantum encoding stage of a QML model with passive steering-based controlled state preparation, which guides the encoded state toward a controlled intermediate state. By tuning the steering strength and the number of steering iterations, the proposed method suppresses the influence of adversarial perturbations while maintaining high clean accuracy and improving adversarial accuracy. Experimental results demonstrate that the passive steering-based defense consistently improves adversarial accuracy across different QML models and datasets under gradient-based adversarial attacks, achieving adversarial accuracy improvements of up to 40.19%.

quant-pharXiv:2605.10954•May 1, 2026•0 citations

Quantum Image Processing

Fei Yan, Salvador Elías Venegas-Andraca

Springer eBooks•Jan 1, 2020•71 citations

Quantum image representation: a review

Marina Lisnichenko, Stanislav Protasov

Quantum Machine Intelligence•Dec 26, 2022•46 citations
The emergence of generative AI poses problems for people in several industries, including artists and other content creators whose work can be trained on without their consent, along with public figures who can be impersonated. As generative AI is now being researched at the quantum level, adversarial AI in quantum remains a largely unexplored topic. With present NISQ architecture and current hardware limitations for quantum image processing in mind, we present a hybrid quantum adversarial AI system using the QPIXL++ library for investigating image perturbations with quantum encoding using Flexible Representation of Quantum Images (FRQI) and Sparse Fast Walsh Hadamard Transform (SFWHT) compression. Topics for investigation in this area include pixel-domain vs frequency-domain perturbations under quantum compression, optimal perturbation strength, quantum superposition, working around quantum hardware limitations and applying optimization techniques with models as DINOv2. This project aims to establish a foundation and launch new research directions in hybrid quantum machine learning and adversarial AI. After completion of the study, it was found that the mean preserved perturbation value for the images was 99.99%, meaning that most of the classically introduced perturbations survived quantum encoding. Additionally, the mean SSIM vaule was close to 1 and the mean PSNR value was 32.30 dB which is above the accepted threshold for indistinguishable images to human eyes.
Quantum machine-learning (QML) models are increasingly considered for security applications, yet their adversarial robustness in malware analysis remains insufficiently understood. This paper presents Q-MALDetect, a hybrid convolutional and variational-quantum detector for byte-image representations of x86-64 Executable and Linkable Format (ELF) binaries, and evaluates it under feature-space and a range of Grad-CAM-guided, executable-aware evasion attacks. The model achieves 97.2% training accuracy, but its robustness deteriorates as attack strength increases, with the strongest evaluated feature-space attack reaching a 78.7% attack success rate (ASR). Among executable-aware attacks, Grad-CAM-guided no-operation (NOP) insertion produces the highest conditional ASR of 32.9%. Dynamic adversarial training is the most effective evaluated defense, reducing feature-space ASR to as low as 5.0%, executable-aware ASR to 7.7%, and the sequential-attack ASR to 34.0%. In contrast, convolutional-autoencoder preprocessing increases executable-aware ASR to 39.7%, coinciding with a rise in Grad-CAM attribution mass within the mutable .text region from 35.2% to 45.7%. Ablation results reveal no monotonic relationship between qubit count and robustness, while the encoding interface modifies the learned attack surface without neutralizing the strongest attack. These findings demonstrate that the effectiveness of attacks and defenses depends strongly on the perturbation constraints and evaluation conditions. Reliable deployment of QML malware detectors therefore requires assessment against both feature-space and executable-aware attacks rather than reliance on clean accuracy or generic denoising alone.
Quantum machine learning (QML) provides a promising framework for leveraging quantum-mechanical effects in learning tasks. However, its vulnerability to adversarial perturbations remains a major challenge for practical deployment. In QML systems, small perturbations applied to classical inputs can propagate through the quantum encoding stage and distort the resulting quantum state, thereby degrading model performance. In this work, we propose a defense mechanism that replaces the conventional quantum encoding stage of a QML model with passive steering-based controlled state preparation, which guides the encoded state toward a controlled intermediate state. By tuning the steering strength and the number of steering iterations, the proposed method suppresses the influence of adversarial perturbations while maintaining high clean accuracy and improving adversarial accuracy. Experimental results demonstrate that the passive steering-based defense consistently improves adversarial accuracy across different QML models and datasets under gradient-based adversarial attacks, achieving adversarial accuracy improvements of up to 40.19%.