Qian Li
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Qian Li has a total of 35 co-authors. Here are the top co-authors by number of shared publications:
Qingyuan Hu
School of Computer Science and Technology, Xian Jiaotong University, Xian, China
5 shared publications- Saiyu Qi
School of Computer Science and Technology, Xian Jiaotong University, Xian, China
5 shared publications - Di Wu
School of Computer Science and Technology, Xian Jiaotong University, Xian, China
4 shared publications - Yong Qi
School of Computer Science and Technology, Xian Jiaotong University, Xian, China
4 shared publications - Chao Shen
Xi'an Jiaotong University, Xi'an, China
3 shared publications - Chenhao Lin
School of Cyber Science and Engineering, Xian Jiaotong University, Xian, China
3 shared publications
Latest Publications
- 4 August 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Dependable and Secure Computing
- Vol. 22 (6) , 7125-7138
AMA: Adaptive Model Poisoning Attacks Towards Federated Learning
Federated Learning (FL) is vulnerable to model poisoning attacks, where malicious updates (e.g., gradients) can adversely interfere with the global model. Existing attacks typically rely heavily on the updates of benign clients and aggregation algorithms to craft malicious updates. However, the benign updates and aggregation algorithms are usually hard to access for attackers, which makes their attacks weak and volatile. Therefore, in this work, we aim to design an adaptive model poisoning attack based on the agnostic adversary. Specifically, we propose a new concept from the perspective of adversarial learning, called adversarial model perturbation. This perturbation targets the parameters of the local model and aims to maximally mislead its predictions. Then, we develop a novel adaptive model poisoning attack named Adversarial Model Attack (AMA), which utilizes the adversarial model perturbation as the malicious updates to attack the global model. Instead of the benign updates and aggregation algorithms, we only leverage the original data of the malicious client to adaptively craft the malicious updates. AMA resolves the conflict between the knowledge requirement of the adversary and the impact of model poisoning attacks. Empirical results against multiple robust FL methods show that AMA surpasses state-of-the-art attack methods and updates the benchmark of attack impact on Fedavg, Trimean, Multi-Krum, FoundationFL, RFA, and Median.
- 4 August 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Dependable and Secure Computing
- Vol. 22 (6) , 7125-7138
AMA: Adaptive Model Poisoning Attacks Towards Federated Learning
Federated Learning (FL) is vulnerable to model poisoning attacks, where malicious updates (e.g., gradients) can adversely interfere with the global model. Existing attacks typically rely heavily on the updates of benign clients and aggregation algorithms to craft malicious updates. However, the benign updates and aggregation algorithms are usually hard to access for attackers, which makes their attacks weak and volatile. Therefore, in this work, we aim to design an adaptive model poisoning attack based on the agnostic adversary. Specifically, we propose a new concept from the perspective of adversarial learning, called adversarial model perturbation. This perturbation targets the parameters of the local model and aims to maximally mislead its predictions. Then, we develop a novel adaptive model poisoning attack named Adversarial Model Attack (AMA), which utilizes the adversarial model perturbation as the malicious updates to attack the global model. Instead of the benign updates and aggregation algorithms, we only leverage the original data of the malicious client to adaptively craft the malicious updates. AMA resolves the conflict between the knowledge requirement of the adversary and the impact of model poisoning attacks. Empirical results against multiple robust FL methods show that AMA surpasses state-of-the-art attack methods and updates the benchmark of attack impact on Fedavg, Trimean, Multi-Krum, FoundationFL, RFA, and Median.
- 4 June 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Information Forensics and Security
- Vol. 20 (15566013) , 6005-6018
Robust Adversarial Defenses in Federated Learning: Exploring the Impact of Data Heterogeneity
Federated Learning (FL) enables geographically distributed clients to collaboratively train machine learning models by exchanging local model parameters while preserving data privacy. In practice, FL faces two critical challenges. First, it is vulnerable to security issues as malicious clients would artificially harm the functionality of FL by launching poisoning attacks. Second, the inherent data heterogeneity among clients (termed Non-IID data in FL) naturally arises from distributed data ownership and significantly degrades model convergence and accuracy. However, with studies separately devoted to these two research lines, the interplay between data heterogeneity and security remains poorly understood. In this paper, we systematically investigate the relationship between data heterogeneity and adversarial robustness in FL. Specifically, we propose novel data partitioning algorithms that simulate Label-Conditional Non-IID and Feature-Conditional Non-IID with quantifiable heterogeneity levels. Further, we conduct extensive experiments to evaluate classical defense methods in the practical FL environment under state-of-the-art untargeted attacks. With results in various settings, we separately analyze the connection between Non-IID to defenses and attacks. Regarding attacks, with similar effects on models, Non-IID impacts the training in a different way compared with attacks. The interaction between attacks and Non-IID provides an opportunity to cause severe damage to FL. Regarding defenses, Non-IID induces heterogeneity in model distribution among clients which raises the difficulty of maintaining fidelity and robustness for defense methods.
- 4 June 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Information Forensics and Security
- Vol. 20 (15566013) , 6005-6018
Robust Adversarial Defenses in Federated Learning: Exploring the Impact of Data Heterogeneity
Federated Learning (FL) enables geographically distributed clients to collaboratively train machine learning models by exchanging local model parameters while preserving data privacy. In practice, FL faces two critical challenges. First, it is vulnerable to security issues as malicious clients would artificially harm the functionality of FL by launching poisoning attacks. Second, the inherent data heterogeneity among clients (termed Non-IID data in FL) naturally arises from distributed data ownership and significantly degrades model convergence and accuracy. However, with studies separately devoted to these two research lines, the interplay between data heterogeneity and security remains poorly understood. In this paper, we systematically investigate the relationship between data heterogeneity and adversarial robustness in FL. Specifically, we propose novel data partitioning algorithms that simulate Label-Conditional Non-IID and Feature-Conditional Non-IID with quantifiable heterogeneity levels. Further, we conduct extensive experiments to evaluate classical defense methods in the practical FL environment under state-of-the-art untargeted attacks. With results in various settings, we separately analyze the connection between Non-IID to defenses and attacks. Regarding attacks, with similar effects on models, Non-IID impacts the training in a different way compared with attacks. The interaction between attacks and Non-IID provides an opportunity to cause severe damage to FL. Regarding defenses, Non-IID induces heterogeneity in model distribution among clients which raises the difficulty of maintaining fidelity and robustness for defense methods.
- 5 June 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Information Forensics and Security
- Vol. 19 (15566013) , 8673-8684
Attention-SA: Exploiting Model-Approximated Data Semantics for Adversarial Attack
Adversarial Defense of deep neural networks have gained significant attention and there have been active research efforts on model blind-points for attacking such as gradient-based attack and pre-defined semantic manipulation. However, they often lack clear adversarial pattern connecting model extracted notion and are restricted to fixed constraint, making the gradual inability to proposed robust defense. In this paper, we propose to utilize the learned semantics of model, possibly not be the true one for the correct prediction, as inspiring clue in adversarial example construction. And we propose a new attention-based semantic oriented adversarial attack without any prior constraint about semantic preservation, dubbed Attention-SA from the learned task-related decision factors perspective. Specifically, to capture the learned factor, we introduce a post-hoc soft attention with a gradient-sensitivity activation consistency to probe the information of latent representation that bridge the input and prediction. With the attention guidance, we perturb the separated and semantic units, then back-propagate the variation onto input to discover expanded adversarial examples. Finally, extensive performance evaluations on CIFAR-10 and ImageNet datasets demonstrate the superiority of our proposed method. And we verify the effectiveness of our method on various robust defenses.
- 5 June 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Information Forensics and Security
- Vol. 19 (15566013) , 8673-8684
Attention-SA: Exploiting Model-Approximated Data Semantics for Adversarial Attack
Adversarial Defense of deep neural networks have gained significant attention and there have been active research efforts on model blind-points for attacking such as gradient-based attack and pre-defined semantic manipulation. However, they often lack clear adversarial pattern connecting model extracted notion and are restricted to fixed constraint, making the gradual inability to proposed robust defense. In this paper, we propose to utilize the learned semantics of model, possibly not be the true one for the correct prediction, as inspiring clue in adversarial example construction. And we propose a new attention-based semantic oriented adversarial attack without any prior constraint about semantic preservation, dubbed Attention-SA from the learned task-related decision factors perspective. Specifically, to capture the learned factor, we introduce a post-hoc soft attention with a gradient-sensitivity activation consistency to probe the information of latent representation that bridge the input and prediction. With the attention guidance, we perturb the separated and semantic units, then back-propagate the variation onto input to discover expanded adversarial examples. Finally, extensive performance evaluations on CIFAR-10 and ImageNet datasets demonstrate the superiority of our proposed method. And we verify the effectiveness of our method on various robust defenses.
- 3 July 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Information Forensics and Security
- Vol. 18 (15566013) , 5936-5949
Revisiting Gradient Regularization: Inject Robust Saliency-Aware Weight Bias for Adversarial Defense
Despite regularizing the Jacobians of neural networks to enhance model robustness has directly theoretical correlation with model prediction stability, a large defense performance gap exists when compared to the empirically perturbation-based adversarial training e.g. PGD-based, which enjoys nice discriminative saliency maps as well. To mitigate this issue, in this paper we first analyze the dilemma that the gradient map of its resulting model has no content hierarchy to mark out salient profile of input, as a negative signal of the obstructive for effective adversarial defense. Based on this, we argue that incorporating robust gradient-based saliency properties into regularized training may be helpful to reduce the performance gap. Specifically, we propose a simple method called Saliency-aware Gradient Regularization (SAGR), where a biased weight distribution strategy is introduced on positive gradient to structure and increase the impact of class-gradient components inside the Jacobian of model. The strategy maintains the dominant role of saliency-critical true-class gradient in learning process and differentiates diverse importance of gradient sensitivities that would localize input salient areas. Herein we interpret the sharpness of true-class sensitivity as robust recognition of more learning-relevant features e.g., regions containing dominant object in image for classification. Instead, false-class parts are considered as recognition-irrelevant nuisance factors e.g. the backgrounds, which are thus depressed with more strength. Experimental results demonstrate the efficacy of the proposed method and validate that distinguishment of sensitivities could further yield more robustness gain and sharper gradient saliency map.
- 3 July 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Information Forensics and Security
- Vol. 18 (15566013) , 5936-5949
Revisiting Gradient Regularization: Inject Robust Saliency-Aware Weight Bias for Adversarial Defense
Despite regularizing the Jacobians of neural networks to enhance model robustness has directly theoretical correlation with model prediction stability, a large defense performance gap exists when compared to the empirically perturbation-based adversarial training e.g. PGD-based, which enjoys nice discriminative saliency maps as well. To mitigate this issue, in this paper we first analyze the dilemma that the gradient map of its resulting model has no content hierarchy to mark out salient profile of input, as a negative signal of the obstructive for effective adversarial defense. Based on this, we argue that incorporating robust gradient-based saliency properties into regularized training may be helpful to reduce the performance gap. Specifically, we propose a simple method called Saliency-aware Gradient Regularization (SAGR), where a biased weight distribution strategy is introduced on positive gradient to structure and increase the impact of class-gradient components inside the Jacobian of model. The strategy maintains the dominant role of saliency-critical true-class gradient in learning process and differentiates diverse importance of gradient sensitivities that would localize input salient areas. Herein we interpret the sharpness of true-class sensitivity as robust recognition of more learning-relevant features e.g., regions containing dominant object in image for classification. Instead, false-class parts are considered as recognition-irrelevant nuisance factors e.g. the backgrounds, which are thus depressed with more strength. Experimental results demonstrate the efficacy of the proposed method and validate that distinguishment of sensitivities could further yield more robustness gain and sharper gradient saliency map.
- 20 October 2022
- Published by Elsevier in Knowledge-Based Systems
- Vol. 259
Understanding and defending against White-box membership inference attack in deep learning
- 20 October 2022
- Published by Elsevier in Knowledge-Based Systems
- Vol. 259
Understanding and defending against White-box membership inference attack in deep learning
- 24 September 2022
- Published by Springer Nature
- p. 405-424
FLMJR: Improving Robustness of Federated Learning via Model Stability
- 24 September 2022
- Published by Springer Nature
- p. 405-424
FLMJR: Improving Robustness of Federated Learning via Model Stability
- 11 March 2022
- Published by Springer Nature in Personal and Ubiquitous Computing
- Vol. 26 (3) , 911
Correction to: Multi-level word features based on CNN for fake news detection in cultural communication
- 11 March 2022
- Published by Springer Nature in Personal and Ubiquitous Computing
- Vol. 26 (3) , 911
Correction to: Multi-level word features based on CNN for fake news detection in cultural communication
- 24 February 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Engineering Management
- Vol. 71 (00189391) , 12433-12447
Truth: A Blockchain-Aided Secure Reputation System With Genuine Feedbacks
E-commerce has changed the way of shopping and played an important role in the world. Therefore, e-commerce platforms require a secure reputation service, which provides genuine reputation of sellers so as to support potential buyers making the right decision while interacting with an e-commerce platform. Unfortunately, most current e-commerce platforms failed to do so because they fully control the reputation service, which make it easy for an adversary to violate the genuineness of the reputation. In this article, we focus on feedback anonymity and authenticity, which are two critical security properties to ensure a genuine reputation of sellers. We present Truth, a blockchain-aided secure reputation system to remove the trustworthiness of the e-commerce platform while ensuring the two security properties. To overcome the usability and security barriers of blockchain systems, Truth adopts a hybrid framework to rely on the centralized e-commerce platform to provide traditional trading services while the blockchain is only used to authenticate the correctness of feedbacks. To decouple the blockchain from the e-commerce platform, we anchor correctness of feedbacks to correctness of cryptographic tokens and leverage the blockchain to solely escrow the tokens. The Truth also proposes a new concept named purchasing endorsed feedback, which is enforced via a collaborative method to prevent injection of fake feedbacks. We have implemented a prototype to demonstrate that the Truth incurs second-level computation overhead and KB-level communication overhead, and achieves hundreds TPS-level thoughts.
- 24 February 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Engineering Management
- Vol. 71 (00189391) , 12433-12447
Truth: A Blockchain-Aided Secure Reputation System With Genuine Feedbacks
E-commerce has changed the way of shopping and played an important role in the world. Therefore, e-commerce platforms require a secure reputation service, which provides genuine reputation of sellers so as to support potential buyers making the right decision while interacting with an e-commerce platform. Unfortunately, most current e-commerce platforms failed to do so because they fully control the reputation service, which make it easy for an adversary to violate the genuineness of the reputation. In this article, we focus on feedback anonymity and authenticity, which are two critical security properties to ensure a genuine reputation of sellers. We present Truth, a blockchain-aided secure reputation system to remove the trustworthiness of the e-commerce platform while ensuring the two security properties. To overcome the usability and security barriers of blockchain systems, Truth adopts a hybrid framework to rely on the centralized e-commerce platform to provide traditional trading services while the blockchain is only used to authenticate the correctness of feedbacks. To decouple the blockchain from the e-commerce platform, we anchor correctness of feedbacks to correctness of cryptographic tokens and leverage the blockchain to solely escrow the tokens. The Truth also proposes a new concept named purchasing endorsed feedback, which is enforced via a collaborative method to prevent injection of fake feedbacks. We have implemented a prototype to demonstrate that the Truth incurs second-level computation overhead and KB-level communication overhead, and achieves hundreds TPS-level thoughts.
- 19 January 2022
- Published by Elsevier in Knowledge-Based Systems
- Vol. 241
Stochastic Ghost Batch for Self-distillation with Dynamic Soft Label
- 19 January 2022
- Published by Elsevier in Knowledge-Based Systems
- Vol. 241
Stochastic Ghost Batch for Self-distillation with Dynamic Soft Label
- 12 October 2021
- Published by Wiley in Transactions in GIS
- Vol. 26 (1) , 362-384
TiGrid: A raster‐based method for time‐interval information representation with geotagged videos as an example
- 12 October 2021
- Published by Wiley in Transactions in GIS
- Vol. 26 (1) , 362-384
TiGrid: A raster‐based method for time‐interval information representation with geotagged videos as an example