Ruoxi Sun
Research Subjects
Ruoxi Sun has a total of 59 co-authors. Here are the top co-authors by number of shared publications:
- Minhui Xue
CSIRO, Adelaide, SA, Australia and Adelaide University, Adelaide, SA, Australia
10 shared publications - Surya Nepal
Data61, CSIRO, Canberra, ACT, Australia
7 shared publications - Minhui Xue
CSIRO and Adelaide University, Adelaide, Australia
3 shared publications - Seyit Camtepe
CSIRO Data61
3 shared publications - Gareth Tyson
The Hong Kong University of Science and Technology (Guangzhou), Guangzhou, China
2 shared publications - Hammond Pearce
University of New South Wales, Sydney, New South Wales, Australia
2 shared publications
Latest Publications
- Open Access
- 14 November 2026
- Published by Association for Computing Machinery (ACM)
- p. 60-71
SoK: The Security-Safety Continuum of Multimodal Foundation Models
- 14 November 2026
- Published by Association for Computing Machinery (ACM)
- p. 60-71
Open AccessSoK: The Security-Safety Continuum of Multimodal Foundation Models
- 24 December 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Information Forensics and Security
- Vol. 21 (15566013) , 357-372
Unfairness Attack and Unified Provable Defense on AI-Powered Internet of Energy
The critical energy infrastructure is undergoing two significant transformations: the rapid increase in renewable distributed energy resources (DER) and the digitalization of the energy sector, collectively shaping what is known as the Internet of Energy (IoE). Artificial intelligence (AI) has become a widely adopted tool for effectively allocating energy and managing sector-related resources, where ensuring fairness is essential. While inherent unfairness in AI systems is well acknowledged, little attention has been given to evaluating this unfairness and its real-world implications within the context of the IoE. In this study, we take a first step to elucidate the unfairness in AI-powered IoE systems induced by malicious users. We introduce Unfairness Score (UScore), a novel metric designed to evaluate the unfairness of machine learning models in real-world IoE scenarios. We then extensively evaluate unfairness attacks using three IoE tabular datasets, demonstrating that AI model fairness can be compromised through data poisoning, whether in centralized learning (CL) or federated learning (FL) settings. Notably, such compromises can occur when malicious users tamper with only a small subset of the data they control. Finally, we propose a novel approach that unifies fairness and differential privacy (DP) by leveraging DP as a provable defense mechanism. This approach provides a universally applicable solution to unfairness attacks, regardless of whether the learning tasks are classification or regression, and is effective in both FL and CL settings. Our contributions represent a significant step in addressing unfairness and privacy concerns in AI-powered IoE systems.
- 24 December 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Information Forensics and Security
- Vol. 21 (15566013) , 357-372
Unfairness Attack and Unified Provable Defense on AI-Powered Internet of Energy
The critical energy infrastructure is undergoing two significant transformations: the rapid increase in renewable distributed energy resources (DER) and the digitalization of the energy sector, collectively shaping what is known as the Internet of Energy (IoE). Artificial intelligence (AI) has become a widely adopted tool for effectively allocating energy and managing sector-related resources, where ensuring fairness is essential. While inherent unfairness in AI systems is well acknowledged, little attention has been given to evaluating this unfairness and its real-world implications within the context of the IoE. In this study, we take a first step to elucidate the unfairness in AI-powered IoE systems induced by malicious users. We introduce Unfairness Score (UScore), a novel metric designed to evaluate the unfairness of machine learning models in real-world IoE scenarios. We then extensively evaluate unfairness attacks using three IoE tabular datasets, demonstrating that AI model fairness can be compromised through data poisoning, whether in centralized learning (CL) or federated learning (FL) settings. Notably, such compromises can occur when malicious users tamper with only a small subset of the data they control. Finally, we propose a novel approach that unifies fairness and differential privacy (DP) by leveraging DP as a provable defense mechanism. This approach provides a universally applicable solution to unfairness attacks, regardless of whether the learning tasks are classification or regression, and is effective in both FL and CL settings. Our contributions represent a significant step in addressing unfairness and privacy concerns in AI-powered IoE systems.
- Open Access
- 19 November 2025
- Published by Association for Computing Machinery (ACM)
- p. 3117-3131
What's Pulling the Strings? Evaluating Integrity and Attribution in AI Training and Inference through Concept Shift
- 19 November 2025
- Published by Association for Computing Machinery (ACM)
- p. 3117-3131
Open AccessWhat's Pulling the Strings? Evaluating Integrity and Attribution in AI Training and Inference through Concept Shift
- 29 October 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Information Forensics and Security
- Vol. 20 (15566013) , 11816-11829
Traceable and Collision-Resilient Differential Privacy
Differential Privacy (DP) is a preeminent technique for data privacy by introducing noise to sensitive information. However, traditional DP mechanisms excessively rely on third parties to ensure traceability, necessitating strong background assumptions that are frequently impractical in real-world scenarios. This reliance makes it difficult to preserve both privacy and traceability. To address these challenges, we propose a novel Traceable and Collision-Resilient Differential Privacy (TCRDP) mechanism. The TCRDP mechanism simultaneously publishes perturbed results and data fingerprints, retaining partial information from the original data in a collision-resilient manner to facilitate future verification. Moreover, the TCRDP mechanism integrates an innovative noise generation process, leveraging hash values and a customized Laplace-like distribution to produce noise. This strategy mitigates the risk of adversaries compromising privacy through enumeration and yields a more concentrated noise distribution with reduced variance. We evaluated the TCRDP mechanism using three datasets: ICUs, Diabetes, and RAHRD, across various query types. The experimental results demonstrated significant improvements in data utility, with the TCRDP mechanism achieving great reductions in Mean Absolute Error (MAE) and Mean Squared Error (MSE) compared to traditional mechanisms. The TCRDP mechanism also maintained lower Accuracy Loss (AL) across different privacy budgets and dataset sizes, highlighting its robustness and scalability. These findings underscore the potential of the TCRDP mechanism to advance privacy-preserving data analysis, offering significant enhancements over existing methods in both accuracy and utility.
- 29 October 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Information Forensics and Security
- Vol. 20 (15566013) , 11816-11829
Traceable and Collision-Resilient Differential Privacy
Differential Privacy (DP) is a preeminent technique for data privacy by introducing noise to sensitive information. However, traditional DP mechanisms excessively rely on third parties to ensure traceability, necessitating strong background assumptions that are frequently impractical in real-world scenarios. This reliance makes it difficult to preserve both privacy and traceability. To address these challenges, we propose a novel Traceable and Collision-Resilient Differential Privacy (TCRDP) mechanism. The TCRDP mechanism simultaneously publishes perturbed results and data fingerprints, retaining partial information from the original data in a collision-resilient manner to facilitate future verification. Moreover, the TCRDP mechanism integrates an innovative noise generation process, leveraging hash values and a customized Laplace-like distribution to produce noise. This strategy mitigates the risk of adversaries compromising privacy through enumeration and yields a more concentrated noise distribution with reduced variance. We evaluated the TCRDP mechanism using three datasets: ICUs, Diabetes, and RAHRD, across various query types. The experimental results demonstrated significant improvements in data utility, with the TCRDP mechanism achieving great reductions in Mean Absolute Error (MAE) and Mean Squared Error (MSE) compared to traditional mechanisms. The TCRDP mechanism also maintained lower Accuracy Loss (AL) across different privacy budgets and dataset sizes, highlighting its robustness and scalability. These findings underscore the potential of the TCRDP mechanism to advance privacy-preserving data analysis, offering significant enhancements over existing methods in both accuracy and utility.
- Top Cited Papers
- 18 June 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Information Forensics and Security
- Vol. 20 (15566013) , 7211-7226
Hardening LLM Fine-Tuning: From Differentially Private Data Selection to Trustworthy Model Quantization
Critical infrastructures are increasingly integrating artificial intelligence (AI) technologies, including large language models (LLMs), into essential systems and services that are vital to societal functioning. Fine-tuning LLMs for specific domain tasks are crucial for their effective deployment in these contexts, but this process must carefully address both privacy and security concerns. Without proper safeguards, such integration can introduce additional risks, such as data leakage during training and diminished model trustworthiness due to the need for model compression to operate within limited bandwidth and computational capacity constraints. In this paper, we propose Hardening LLM Fine-tuning framework (HARDLLM), which addresses these challenges through two key components: (i) we develop a differentially private data selection method that ensures privacy protection by training the model exclusively on sampled and synthesized public data, thereby preventing any direct use of private data and enhancing leakage resilience throughout the training process, and (ii) we introduce a trustworthiness-aware model quantization approach to improve LLMs performance, such as reducing toxicity, enhancing adversarial robustness, and mitigating stereotypes, while maintaining negligible impact on model utility. Experimental results show that, the proposed algorithm ensures differential privacy when privacy budget is set at ϵ = 0.5, with only a 1% drop in accuracy, while other state-of-the-art methods experience an accuracy drop of at least 20% under the same privacy budget. Additionally, our quantization approach improves the trustworthiness of fine-tuned LLMs by an average of 3-4%, with only a negligible utility loss (approximately 1%) at a 50% compression rate.
- 18 June 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Information Forensics and Security
- Vol. 20 (15566013) , 7211-7226
Top Cited PapersHardening LLM Fine-Tuning: From Differentially Private Data Selection to Trustworthy Model Quantization
Critical infrastructures are increasingly integrating artificial intelligence (AI) technologies, including large language models (LLMs), into essential systems and services that are vital to societal functioning. Fine-tuning LLMs for specific domain tasks are crucial for their effective deployment in these contexts, but this process must carefully address both privacy and security concerns. Without proper safeguards, such integration can introduce additional risks, such as data leakage during training and diminished model trustworthiness due to the need for model compression to operate within limited bandwidth and computational capacity constraints. In this paper, we propose Hardening LLM Fine-tuning framework (HARDLLM), which addresses these challenges through two key components: (i) we develop a differentially private data selection method that ensures privacy protection by training the model exclusively on sampled and synthesized public data, thereby preventing any direct use of private data and enhancing leakage resilience throughout the training process, and (ii) we introduce a trustworthiness-aware model quantization approach to improve LLMs performance, such as reducing toxicity, enhancing adversarial robustness, and mitigating stereotypes, while maintaining negligible impact on model utility. Experimental results show that, the proposed algorithm ensures differential privacy when privacy budget is set at ϵ = 0.5, with only a 1% drop in accuracy, while other state-of-the-art methods experience an accuracy drop of at least 20% under the same privacy budget. Additionally, our quantization approach improves the trustworthiness of fine-tuned LLMs by an average of 3-4%, with only a negligible utility loss (approximately 1%) at a 50% compression rate.
- Open Access
- 22 April 2025
- Published by Association for Computing Machinery (ACM)
- p. 978-989
50 Shades of Deceptive Patterns: A Unified Taxonomy, Multimodal Detection, and Security Implications
- 22 April 2025
- Published by Association for Computing Machinery (ACM)
- p. 978-989
Open Access50 Shades of Deceptive Patterns: A Unified Taxonomy, Multimodal Detection, and Security Implications
- 30 January 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Dependable and Secure Computing
- Vol. 22 (4) , 3661-3675
Leakage-Resilient and Carbon-Neutral Aggregation Featuring the Federated AI-Enabled Critical Infrastructure
AI-enabled critical infrastructures (ACIs) integrate artificial intelligence (AI) technologies into various essential systems and services that are vital to the functioning of society, offering significant implications for efficiency, security and resilience. While adopting decentralized AI approaches (such as federated learning technology) in ACIs is plausible, private and sensitive data are still susceptible to data reconstruction attacks through gradient optimization. In this work, we propose Compressed Differentially Private Aggregation (CDPA), a leakage-resilient, communication-efficient, and carbon-neutral approach for ACI networks. Specifically, CDPA has introduced a novel random bit-flipping mechanism as its primary innovation. This mechanism first converts gradients into a specific binary representation and then selectively flips masked bits with a certain probability. The proposed bit-flipping introduces a larger variance to the noise while providing differentially private protection and commendable efforts in energy savings while applying vector quantization techniques within the context of federated learning. The experimental evaluation indicates that CDPA can reduce communication cost by half while preserving model utility. Moreover, we demonstrate that CDPA can effectively defend against state-of-the-art data reconstruction attacks in both computer vision and natural language processing tasks. We highlight existing benchmarks that generate 2.6× to over 100× more carbon emissions than CDPA. We hope that the CDPA developed in this paper can inform the federated AI-enabled critical infrastructure of a more balanced trade-off between utility and privacy, resilience protection, as well as a better carbon offset with less communication overhead.
- 30 January 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Dependable and Secure Computing
- Vol. 22 (4) , 3661-3675
Leakage-Resilient and Carbon-Neutral Aggregation Featuring the Federated AI-Enabled Critical Infrastructure
AI-enabled critical infrastructures (ACIs) integrate artificial intelligence (AI) technologies into various essential systems and services that are vital to the functioning of society, offering significant implications for efficiency, security and resilience. While adopting decentralized AI approaches (such as federated learning technology) in ACIs is plausible, private and sensitive data are still susceptible to data reconstruction attacks through gradient optimization. In this work, we propose Compressed Differentially Private Aggregation (CDPA), a leakage-resilient, communication-efficient, and carbon-neutral approach for ACI networks. Specifically, CDPA has introduced a novel random bit-flipping mechanism as its primary innovation. This mechanism first converts gradients into a specific binary representation and then selectively flips masked bits with a certain probability. The proposed bit-flipping introduces a larger variance to the noise while providing differentially private protection and commendable efforts in energy savings while applying vector quantization techniques within the context of federated learning. The experimental evaluation indicates that CDPA can reduce communication cost by half while preserving model utility. Moreover, we demonstrate that CDPA can effectively defend against state-of-the-art data reconstruction attacks in both computer vision and natural language processing tasks. We highlight existing benchmarks that generate 2.6× to over 100× more carbon emissions than CDPA. We hope that the CDPA developed in this paper can inform the federated AI-enabled critical infrastructure of a more balanced trade-off between utility and privacy, resilience protection, as well as a better carbon offset with less communication overhead.
- Open Access
- 30 November 2024
- Published by Association for Computing Machinery (ACM)
- p. 223-231
Privacy in Motion: Implementing Differential Privacy for User Motion in VR
- 30 November 2024
- Published by Association for Computing Machinery (ACM)
- p. 223-231
Open AccessPrivacy in Motion: Implementing Differential Privacy for User Motion in VR
- 17 October 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Dependable and Secure Computing
- Vol. 22 (3) , 2243-2257
On Security Weaknesses and Vulnerabilities in Deep Learning Systems
The security guarantee of AI-enabled software systems (particularly using deep learning techniques as a functional core) is pivotal against the adversarial attacks exploiting software vulnerabilities. However, little attention has been paid to a systematic investigation of vulnerabilities in such systems. A common situation learned from the open source software community is that deep learning engineers frequently integrate off-the-shelf or open-source learning frameworks into their ecosystems. In this work, we specifically look into deep learning (DL) framework and perform the first systematic study of vulnerabilities in DL systems through a comprehensive analysis of identified vulnerabilities from Common Vulnerabilities and Exposures (CVE) and open-source DL tools, including TensorFlow, Caffe, OpenCV, Keras, and PyTorch. We propose a two-stream data analysis framework to explore vulnerability patterns from various databases. We investigate the unique DL frameworks and libraries development ecosystems that appear to be decentralized and fragmented. By revisiting the Common Weakness Enumeration (CWE) List, which provides the traditional software vulnerability related practices, we observed that it is more challenging to detect and fix the vulnerabilities throughout the DL systems lifecycle. Moreover, we conducted a large-scale empirical study of DL vulnerabilities to better understand the patterns of vulnerability and the challenges in fixing them. We have released the full replication package at https://github.com/codelzz/Vulnerabilities4DLSystem . We anticipate that our study can advance the development of secure DL systems.
- 17 October 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Dependable and Secure Computing
- Vol. 22 (3) , 2243-2257
On Security Weaknesses and Vulnerabilities in Deep Learning Systems
The security guarantee of AI-enabled software systems (particularly using deep learning techniques as a functional core) is pivotal against the adversarial attacks exploiting software vulnerabilities. However, little attention has been paid to a systematic investigation of vulnerabilities in such systems. A common situation learned from the open source software community is that deep learning engineers frequently integrate off-the-shelf or open-source learning frameworks into their ecosystems. In this work, we specifically look into deep learning (DL) framework and perform the first systematic study of vulnerabilities in DL systems through a comprehensive analysis of identified vulnerabilities from Common Vulnerabilities and Exposures (CVE) and open-source DL tools, including TensorFlow, Caffe, OpenCV, Keras, and PyTorch. We propose a two-stream data analysis framework to explore vulnerability patterns from various databases. We investigate the unique DL frameworks and libraries development ecosystems that appear to be decentralized and fragmented. By revisiting the Common Weakness Enumeration (CWE) List, which provides the traditional software vulnerability related practices, we observed that it is more challenging to detect and fix the vulnerabilities throughout the DL systems lifecycle. Moreover, we conducted a large-scale empirical study of DL vulnerabilities to better understand the patterns of vulnerability and the challenges in fixing them. We have released the full replication package at https://github.com/codelzz/Vulnerabilities4DLSystem . We anticipate that our study can advance the development of secure DL systems.
- 2 October 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Dependable and Secure Computing
- Vol. 22 (2) , 1827-1844
Iterative Window Mean Filter: Thwarting Diffusion-Based Adversarial Purification
Face authentication systems have brought significant convenience and advanced developments, yet they have become unreliable due to their sensitivity to inconspicuous perturbations, such as adversarial attacks. Existing defenses often exhibit weaknesses when facing various attack algorithms and adaptive attacks or compromise accuracy for enhanced security. To address these challenges, we have developed a novel and highly efficient non-deep-learning-based image filter called the Iterative Window Mean Filter (IWMF) and proposed a new framework for adversarial purification, named IWMF-Diff, which integrates IWMF and denoising diffusion models. These methods can function as pre-processing modules to eliminate adversarial perturbations without necessitating further modifications or retraining of the target system. We demonstrate that our proposed methodologies fulfill four critical requirements: preserved accuracy, improved security, generalizability to various threats in different settings, and better resistance to adaptive attacks. This performance surpasses that of the state-of-the-art adversarial purification method, DiffPure. Our code is released at https://github.com/azrealwang/iwmfdiff .
- 2 October 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Dependable and Secure Computing
- Vol. 22 (2) , 1827-1844
Iterative Window Mean Filter: Thwarting Diffusion-Based Adversarial Purification
Face authentication systems have brought significant convenience and advanced developments, yet they have become unreliable due to their sensitivity to inconspicuous perturbations, such as adversarial attacks. Existing defenses often exhibit weaknesses when facing various attack algorithms and adaptive attacks or compromise accuracy for enhanced security. To address these challenges, we have developed a novel and highly efficient non-deep-learning-based image filter called the Iterative Window Mean Filter (IWMF) and proposed a new framework for adversarial purification, named IWMF-Diff, which integrates IWMF and denoising diffusion models. These methods can function as pre-processing modules to eliminate adversarial perturbations without necessitating further modifications or retraining of the target system. We demonstrate that our proposed methodologies fulfill four critical requirements: preserved accuracy, improved security, generalizability to various threats in different settings, and better resistance to adaptive attacks. This performance surpasses that of the state-of-the-art adversarial purification method, DiffPure. Our code is released at https://github.com/azrealwang/iwmfdiff .