Yihong Zhang
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Yihong Zhang has a total of 10 co-authors. Here are the top co-authors by number of shared publications:
- Guojia Li
School of Cyber Science and Engineering, Zhengzhou University, Zhengzhou 450002, China
3 shared publications - Mingyue Cao
School of Cyber Science and Engineering, Zhengzhou University, Zhengzhou 450002, China
3 shared publications - Yan Cao
Zhengzhou University
3 shared publications - Simin Xu
School of Cyber Science and Engineering, Zhengzhou University, Zhengzhou 450002, China
2 shared publications - Chengyao Xue
School of Cyber Science and Engineering, Zhengzhou University, Zhengzhou 450002, China
1 shared publication - Da Xiao
College of Engineering, Xi’an International University, No. 18 Yudou Road, Yanta District, Xi’an 710077, China
1 shared publication
Latest Publications
- Open Access
- 5 August 2026
- Published by MDPI AG in Electronics
- Vol. 15 (15) , 3468
SGC: Soft Gradient Collaboration for Backdoor Attacks in Self-Supervised Distillation
Self-supervised knowledge distillation is widely used to compress reusable encoders, but an untrusted distillation implementation can itself become an attack surface. We study an algorithm-level threat in which the teacher encoder and user-visible distillation dataset remain unchanged, while malicious code internally generates trigger-bearing views and optimizes an additional backdoor objective. To instantiate this threat, we propose soft gradient collaboration (SGC), which combines distribution-alignment-based distillation, target-representation-based backdoor design, and conflict-avoidance gradient collaboration to reduce interference with benign representation transfer while embedding a trigger-to-target association in the student encoder. Experiments on CIFAR-10 and STL-10 show that SGC maintains competitive downstream accuracy and effective non-target attack success. Quantitative CKA, feature-distribution, and class-structure analyses further indicate that SGC retains clean representations closer to benign distillation than fixed scalarization or removal of distribution alignment. Its no-defense attack success is not the highest among the compared attacks; instead, its main empirical advantage is stronger residual attack persistence after MIMIC, MKD, and SSLDefender. Under SSLDefender, SGC retains 9.12% non-target ASR on CIFAR-10 and 9.06% on STL-10, the highest residual values among the compared attacks. Additional experiments with a compact ResNet-18 student, multiple target classes and trigger configurations, and a supplemental CIFAR-100 setting broaden the empirical evaluation across student capacity, target semantics, trigger configurations, and label-space complexity. These results show that security assessment of self-supervised distillation should include executable training logic in addition to model weights and visible data. The concealment considered here is limited to dataset-only inspection and clean-output validation; SGC is not claimed to evade source-code auditing, runtime data-flow monitoring, or training-log inspection.
- 5 August 2026
- Published by MDPI AG in Electronics
- Vol. 15 (15) , 3468
Open AccessSGC: Soft Gradient Collaboration for Backdoor Attacks in Self-Supervised Distillation
Self-supervised knowledge distillation is widely used to compress reusable encoders, but an untrusted distillation implementation can itself become an attack surface. We study an algorithm-level threat in which the teacher encoder and user-visible distillation dataset remain unchanged, while malicious code internally generates trigger-bearing views and optimizes an additional backdoor objective. To instantiate this threat, we propose soft gradient collaboration (SGC), which combines distribution-alignment-based distillation, target-representation-based backdoor design, and conflict-avoidance gradient collaboration to reduce interference with benign representation transfer while embedding a trigger-to-target association in the student encoder. Experiments on CIFAR-10 and STL-10 show that SGC maintains competitive downstream accuracy and effective non-target attack success. Quantitative CKA, feature-distribution, and class-structure analyses further indicate that SGC retains clean representations closer to benign distillation than fixed scalarization or removal of distribution alignment. Its no-defense attack success is not the highest among the compared attacks; instead, its main empirical advantage is stronger residual attack persistence after MIMIC, MKD, and SSLDefender. Under SSLDefender, SGC retains 9.12% non-target ASR on CIFAR-10 and 9.06% on STL-10, the highest residual values among the compared attacks. Additional experiments with a compact ResNet-18 student, multiple target classes and trigger configurations, and a supplemental CIFAR-100 setting broaden the empirical evaluation across student capacity, target semantics, trigger configurations, and label-space complexity. These results show that security assessment of self-supervised distillation should include executable training logic in addition to model weights and visible data. The concealment considered here is limited to dataset-only inspection and clean-output validation; SGC is not claimed to evade source-code auditing, runtime data-flow monitoring, or training-log inspection.
- Open Access
- 29 October 2025
- Published by MDPI AG in Electronics
- Vol. 14 (21) , 4236
From 2D-Patch to 3D-Camouflage: A Review of Physical Adversarial Attack in Object Detection
Deep neural networks have demonstrated remarkable performance in object detection tasks; however, they remain highly susceptible to adversarial attacks. Previous surveys in computer vision have provided considerable coverage of physical adversarial attacks, yet the aspect of systematically categorizing and evaluating their physical deployment methods for object detection has not received commensurate focus. To address this gap, we categorize physical adversarial attacks into three primary classes based on the physical deployment of adversarial patterns: manipulating 2D physical objects, injecting adversarial signals, and placing 3D adversarial camouflage. These categories are further analyzed and compared across nine key attributes. Furthermore, we elucidate the relationship between physical adversarial attacks against object detection models and two critical properties: transferability and perceptibility. Our findings indicate that while attacks involving the manipulation of 2D physical objects are relatively straightforward to deploy, their adversarial patterns are often perceptible to human observers. Similarly, 3D adversarial camouflage tends to lack stealthiness, whereas adversarial signal injection offers stronger imperceptibility. However, all three attack types exhibit limited transferability across different models and modalities. Finally, we discuss current challenges and propose actionable directions for future research, aiming to foster the development of more robust object detection systems.
- 29 October 2025
- Published by MDPI AG in Electronics
- Vol. 14 (21) , 4236
Open AccessFrom 2D-Patch to 3D-Camouflage: A Review of Physical Adversarial Attack in Object Detection
Deep neural networks have demonstrated remarkable performance in object detection tasks; however, they remain highly susceptible to adversarial attacks. Previous surveys in computer vision have provided considerable coverage of physical adversarial attacks, yet the aspect of systematically categorizing and evaluating their physical deployment methods for object detection has not received commensurate focus. To address this gap, we categorize physical adversarial attacks into three primary classes based on the physical deployment of adversarial patterns: manipulating 2D physical objects, injecting adversarial signals, and placing 3D adversarial camouflage. These categories are further analyzed and compared across nine key attributes. Furthermore, we elucidate the relationship between physical adversarial attacks against object detection models and two critical properties: transferability and perceptibility. Our findings indicate that while attacks involving the manipulation of 2D physical objects are relatively straightforward to deploy, their adversarial patterns are often perceptible to human observers. Similarly, 3D adversarial camouflage tends to lack stealthiness, whereas adversarial signal injection offers stronger imperceptibility. However, all three attack types exhibit limited transferability across different models and modalities. Finally, we discuss current challenges and propose actionable directions for future research, aiming to foster the development of more robust object detection systems.
- Open Access
- 23 September 2025
- Published by MDPI AG in Electronics
- Vol. 14 (19) , 3753
Invisible Backdoor Attack Based on Dual-Frequency- Domain Transformation
Backdoor attacks are recognized as a significant security threat to deep learning. Such attacks can induce models to perform abnormally with inputs that contain predefined triggers, while maintaining state-of-the-art (SOTA) performance on clean data. Research indicates that existing backdoor attacks in the spatial domain have the problems of poor stealthiness and limited effectiveness. Based on the dispersion of adding perturbations in the frequency domain and the idea that multiple frequency-domain transformations can achieve different levels of feature fusion, we propose a dual-frequency-domain transformation backdoor attack method called DFDT (dual-frequency-domain transformation). DFDT executes dual-frequency-domain transformation on both clean samples and a trigger image, then conducts feature fusion in the frequency domain to augment the stealthiness of the poisoned samples. In addition, we introduce regularization samples to reduce the latent separability of clean and poisoned samples. We thoroughly evaluate the DFDT on three image datasets: CIFAR-10, GTSRB, and CIFAR-100. The experimental results show that the DFDT achieves greater stealthiness and effectiveness, achieving an attack success rate (ASR) that approximates 100% and a benign accuracy (BA) nearing 94%. Furthermore, we illustrate that DFDT can successfully evade state-of-the-art defenses, including STRIP, NC, and I-BAU.
- 23 September 2025
- Published by MDPI AG in Electronics
- Vol. 14 (19) , 3753
Open AccessInvisible Backdoor Attack Based on Dual-Frequency- Domain Transformation
Backdoor attacks are recognized as a significant security threat to deep learning. Such attacks can induce models to perform abnormally with inputs that contain predefined triggers, while maintaining state-of-the-art (SOTA) performance on clean data. Research indicates that existing backdoor attacks in the spatial domain have the problems of poor stealthiness and limited effectiveness. Based on the dispersion of adding perturbations in the frequency domain and the idea that multiple frequency-domain transformations can achieve different levels of feature fusion, we propose a dual-frequency-domain transformation backdoor attack method called DFDT (dual-frequency-domain transformation). DFDT executes dual-frequency-domain transformation on both clean samples and a trigger image, then conducts feature fusion in the frequency domain to augment the stealthiness of the poisoned samples. In addition, we introduce regularization samples to reduce the latent separability of clean and poisoned samples. We thoroughly evaluate the DFDT on three image datasets: CIFAR-10, GTSRB, and CIFAR-100. The experimental results show that the DFDT achieves greater stealthiness and effectiveness, achieving an attack success rate (ASR) that approximates 100% and a benign accuracy (BA) nearing 94%. Furthermore, we illustrate that DFDT can successfully evade state-of-the-art defenses, including STRIP, NC, and I-BAU.
- Open Access
- 6 July 2025
- Published by MDPI AG in Electronics
- Vol. 14 (13) , 2724
BASK: Backdoor Attack for Self-Supervised Encoders with Knowledge Distillation Survivability
Backdoor attacks in self-supervised learning pose an increasing threat. Recent studies have shown that knowledge distillation can mitigate these attacks by altering feature representations. In response, we propose BASK, a novel backdoor attack that remains effective after distillation. BASK uses feature weighting and representation alignment strategies to implant persistent backdoors into the encoder’s feature space. This enables transferability to student models. We evaluated BASK on the CIFAR-10 and STL-10 datasets and compared it with existing self-supervised backdoor attacks under four advanced defenses: SEED, MKD, Neural Cleanse, and MiMiC. Our experimental results demonstrate that BASK maintains high attack success rates while preserving downstream task performance. This highlights the robustness of BASK and the limitations of current defense mechanisms.
- 6 July 2025
- Published by MDPI AG in Electronics
- Vol. 14 (13) , 2724
Open AccessBASK: Backdoor Attack for Self-Supervised Encoders with Knowledge Distillation Survivability
Backdoor attacks in self-supervised learning pose an increasing threat. Recent studies have shown that knowledge distillation can mitigate these attacks by altering feature representations. In response, we propose BASK, a novel backdoor attack that remains effective after distillation. BASK uses feature weighting and representation alignment strategies to implant persistent backdoors into the encoder’s feature space. This enables transferability to student models. We evaluated BASK on the CIFAR-10 and STL-10 datasets and compared it with existing self-supervised backdoor attacks under four advanced defenses: SEED, MKD, Neural Cleanse, and MiMiC. Our experimental results demonstrate that BASK maintains high attack success rates while preserving downstream task performance. This highlights the robustness of BASK and the limitations of current defense mechanisms.