My research interests focus on Machine Learning (ML) and Federated Learning (FL).

🔥 News

• 2025.03:  🎉🎉 Two papers on vulnerability disclosure on T2I models and vunerability exploration in CFL are accepted by Oakland’25!
• 2024.12:  🎉🎉 I am invited to be an AE reviewer of USENIX Security’25!
• 2024.12:  🎉🎉 I am invited to be a reviewer of IEEE TDSC, TIFS, TSC, and Neural Network!
• 2024.10:  🎉🎉 I am invited to be a reviewer of PAKDD’25!
• 2024.08:  🎉🎉 Our paper on privacy leakage of language models is accepted by CCS’24!
• 2024.07:  🎉🎉 I am invited to be a reviewer of ACM CIKM’24!
• 2024.07:  🎉🎉 Our paper on unveiling IP vulnerabilities of GAN-based distributed ML is accepted by CIKM’24!
• 2023.08:  🎉🎉 Our paper on formalizing neural network perturbation is accepted by ICFEM’23!
• 2023.06:  🎉🎉 I am invited to be an AE reviewer of NDSS’24!
• 2023.06:  🎉🎉 I pass my first-year progress review confirmation!
• 2023.05:  🎉🎉 Our paper on mitigating membership inference threats in FL is accepted by AsiaCCS’23!
• 2022.04:  🎉🎉 I become a Ph.D. student at the University of Queensland!

📝 Selected Publications

S&P'25

Modifier Unlocked: Jailbreaking Text-to-Image Models Through Prompts

Shuofeng Liu, Mengyao Ma, Minhui Xue, Guangdong Bai. IEEE Symposium on Security and Privacy (Oakland), 2025.

S&P'25

Practical Poisoning Attacks with Limited Byzantine Clients in Clustered Federated Learning

Viet Vo, Mengyao Ma, Guangdong Bai, Ryan Ko, Surya Nepal. IEEE Symposium on Security and Privacy (Oakland), 2025.

CIKM'24

Unveiling Intellectual Property Vulnerabilities of GAN-Based Distributed Machine Learning through Model Extraction Attacks

Mengyao Ma, Shuofeng Liu, Mahawaga Arachchige Pathum Chamikara, Mohan Baruwal Chhetri, Guangdong Bai. ACM International Conference on Information and Knowledge Management (CIKM), Jul. 2024.

CCS'24

Uncovering Gradient Inversion Risks in Practical Language Model Training

Xinguo Feng, Zhongkui Ma, Zihan Wang, Chegne Eu Joe, Mengyao Ma, Alsharif Abuadbba, Guangdong Bai. ACM Computer and Communications Security Conference (CCS), Aug. 2024.

AsiaCCS'23

LoDen: Making Every Client in Federated Learning a Defender Against the Poisoning Membership Inference Attacks

Mengyao Ma, Yanjun Zhang, Pathum Chamikara Mahawaga Arachchige, Leo Yu Zhang, Mohan Baruwal Chhetri, Guangdong Bai. ACM ASIA Conference on Computer and Communications Security (AsiaCCS), Apr. 2023.

ICFEM'23

Formalizing Robustness Against Character-Level Perturbations for Neural Network Language Models

Zhongkui Ma, Xinguo Feng, Zihan Wang, Shuofeng Liu, Mengyao Ma, Hao Guan, Mark Huasong Meng. International Conference on Formal Engineering Methods (ICFEM), Aug. 2023.

🎖 Awards

• 2024.09: 🏆 SIGIR Travel Grant for presenting at CIKM’24, Boise, USA.
• 2022.01: 🎓 CSIRO’s Data61 Full Scholarship with top-up funding.
• 2021.11: 🏅 Dean’s Commendation for Academic Excellence at UQ.

📌 Service

• Web Chair: IWQoS 2025.
• Leader of Local Committee: ICFEM 2023.
• AE Reviewer: USENIX Security 2025, NDSS 2024.
• Reviewer: PAKDD 2025, IEEE TDSC, TSC, TIFS, Neural Networks.