News Feed May, June, and July 2024

Accepted Papers

  • Xiangyu Hui, Sid Chi-Kin Chau. “LLRing: Logarithmic Linkable Ring Signatures with Transparent Setup”, accepted by ESORICS 2024 (CORE A, Data61 target venue). The paper presents an attack on a linkable ring signature scheme, DualDory (ESORICS ‘22), from IBM Research. Additionally, it proposes new logarithmic linkable ring signature schemes with transparent setup, which is more efficient in terms of verification, under both Dlog and bilinear pairing settings.
  • Kane Walter, Surya Nepal, Salil Kanhere. “Exploiting Layerwise Feature Representation Similarity For Backdoor Defence in Federated Learning”, accepted by ESORICS 2024 (CORE A, Data61 target venue). The paper proposes a defence against backdoor attacks on Federated Learning systems. The defence frames backdoor attacks as a similarity measurement task in the feature representation space of neural networks and leverages Centered Kernel Alignment (Kornblith et al., 2019) to detect malicious model updates.
  • Mohan Baruwal Chhetri, Shahroz Tariq, Ronal Singh, Fatemeh Jalalvand, Cecile Paris, Surya Nepal, “Towards Human-AI Teaming to Mitigate Alert Fatigue in Security Operations Centres”, accepted by ACM Transactions on Internet Technology. This position paper presents a vision and proposal for harnessing the power of human-AI teaming to enhance the efficiency and effectiveness of Security Operations Centres (SOC) operations. It also presents a conceptual framework for human-AI teaming, termed A2C Framework to support three distinct modes of decision-making in human-AI teams: automated, augmented and collaborative.
  • Diksha Goel, Kristen Moore, Mingyu Guo, Derek Wang, Minjune Kim, Seyit Camtepe, “Optimizing Cyber Defense in Dynamic Active Directories through Reinforcement Learning”, ESORICS 2024 (CORE A, Data61 target venue). This paper proposes dual reinforcement learning-based policies for advancing cyber defense in dynamic and large-scale environments. Our results demonstrate that the proposed approach accurately models attacker behavior and generates highly effective defense.
  • Bao Gia Doan, Dang Quang Nguyen, Paul Montague, Tamas Abraham, Olivier De Vel, Seyit Camtepe, Salil S. Kanhere, Ehsan Abbasnejad and Damith C. Ranasinghe, Detecting Adversarial Malware For Free, ESORICS 2024 (CORE A, Data61 target venue).
  • Sebastian Kish, C. Thapa, M. Sayat, H. Suzuki, J. Pieprzyk, and S. Camtepe, Mitigation of Channel Tampering Attacks in Continuous-Variable Quantum Key Distribution, Physical Review Research, 2024. This paper introduces the channel amplification (CA) attack in continuous-variable quantum key distribution (CV-QKD) and proposes a machine learning-based detection and postselection strategy. These methods improve the secret key rate (SKR) and enhance CV-QKD robustness against CA and hybrid CA-DoS attacks.
  • Hoang Pham, The-Anh Ta, Anh Tran, Khoa Doan, “Flatness-aware Sequential Learning Generates Resilient Backdoors”, accepted at the 18th European Conference on Computer Vision (ECCV 2024, CORE Rank A*). This paper presents a new backdoor attack on deep learning model that can bypass the state-of-the-art finetuning backdoor defenses.
  • Trina Chang , Zhibo Hu, Xiaoyu Li, Shuiqiao Yang, Jiaojiao Jiang and Nan Sun, DiHAN: A Novel Dynamic Hierarchical Graph Attention Network for Fake News Detection, accepted at CIKM 2024 (33rd ACM International Conference on Information and Knowledge Management (CIKM 2024, CORE Rank A). This paper proposes a novel GNN based fake news detection algorithm for social media corpus.
  • Rayne Holland, Chandra Thapa, Wei Shao, Seyit Camtepe and Sarah Ali Siddiqui, “One-shot Collaborative Data Distillation,” accepted at ECAI 2024 (core rank A conference). This paper introduces the first collaborative data distillation technique, called CollabDM, which captures the global distribution of the data and requires only a single round of communication between client and server (reduce the communication cost of data sharing for ML and ML-based applications). The proposed method outperforms the state-of-the-art one-shot learning method on skewed data in distributed learning environments.  
  • Mengyao Ma, Shuofeng Liu, M. A.P.Chamikara, Mohan Baruwal Chhetri, Guangdong Bai, “Unveiling Intellectual Property Vulnerabilities of GAN-Based Distributed Machine Learning through Model Extraction Attacks”, accepted at CIKM 2024 (Core A). This paper introduces a novel attack, MoEx, targeting the security of Generative Adversarial Networks (GANs) in distributed learning environments. By analysing data exchanged during training, MoEx effectively extracts the GAN generator model, posing a significant threat to intellectual property protection. 
  • Fei Zhu, Guo, Duan, Sharif Abuadbba, Yi, Xun, Luo, Junwei, Kumari, Saru, Peng, Tao, “Lightweight Verifiable Privacy-Preserving Data Aggregation for Smart Grids” accepted by IEEE IoT 2024. This paper was part of Fei Zhu Data61 funded PhD work and introduced a new lightweight data aggregation scheme for smart grids called LVP2DA, which enhances security and reduces costs. It avoids complex cryptographic operations, making it suitable for smart meters with limited resources. The scheme uses a certificate-based approach without pairings, which has proven to be secure. 

Projects

  • Mohan Baruwal Chhetri, Minjune Kim, and Lauren Ferro conducted a focus group session for the CSCRC project CyberBUDDY on 16 July 2024. This was the first outing of the project. The CyberBUDDY platform supports SMEs on their cyber maturity journey by using state of the art learning, gamification and nudging techniques. It motivates SMEs to take a do-something approach instead of relying on consultants and interventions. 
  • Tooba Aamir presented the Mission CIPR Report to DHA on the interdependencies and impact of Critical Infrastructure Sectors. The report contains the findings of the workshop conducted by the DHA’s Resilience Expert Advisory Group (REAG) via the Trusted Information Sharing Network (TISN). More details available here.
  • A large contingent from the CQS Group attended the last CSCRC Symposium in Canberra on 20-21 June 2025. Our group is leading several projects funded through the CSCRC including ACDC, Corporates Compromised, CyberBUDDY, OptimShare, SOCRATES etc.

Awards

  • The CQS group had a clean sweep at the Victorian iAwards 2024. Quantum Shield for Artificial Intelligence Systems won the Technology Platform category. Muhammad Usman and his team designed and implemented novel quantum artificial intelligence systems that were extremely robust against attacks. The work established the integration of quantum into artificial intelligence that could lead the world to a new era of trust and reliability in autonomous systems. Read more about Quantum Shield here. Similarly, OptimShare took home the Merit Award in the Technology Platform category. OptimShare facilitates privacy-preserving data sharing between government and private sectors. It is envisioned it could become the standard for privacy-preserving data sharing. Learn more about OptimShare here. The Project Team includes M. A. P. Chamikara, Seung Jang, Dongxi Liu, Seyit Camtepe, and Surya Nepal.

  • TAPE was a finalist at the iAwards NSW 2024. TAPE is an AI-powered tool that collaborates with Security analysts to make sense of a large number of email alerts (thousands on a daily basis), identify malicious campaigns, and prioritise them. The TAPE Project Team includes Sharif Abuadbba, Seung Jang, Wei Kang, Shuo Wang, and Nan Wang. 
  • REDWIRE was a finalist at the iAwards NSW 2024. REDWIRE is a novel tool based on a patented algorithm to prevent ransomware or unauthorised insider encryption of corporate files shared through cloud/network storage. It is the only solution of its type which can stop ransomware encryption regardless of ransomware or end-point types. The team includes Arash Mahboubi (CSU), Seyit Camtepe, and Josef Pieprzyk.

Staff and Students

  • We farewelled Benham Tonekaboni, Wei Kang and Geoff Jarrad.
  • Kane Walters officially graduated his PhD Degree from UNSW.
  • Sid Chau and Seung Jang were appointed as supervisors of the Systems Security and Emerging Tech Cybersecurity sub-teams.
  • Ejaz Ahmed received the Julius Career Award.
  • Mohan Baruwal Chhetri was elevated to IEEE Senior Member.

Networking / Conferences

  • Surya Nepal participated in the Dagstuhl Seminar on Security and Privacy of Current and Emerging IoT Devices and Systems from Jul 28 – Aug 02, 2024. He also visited the University of Newcastle for the UKRI Project on Internet of Energy Security.

  • Eromanga Adermann was a keynote speaker for the Women in AI APAC Summit 2024, which was held in Sydney from June 27-28, 2024. She gave a talk introducing Quantum Machine Learning to an audience of largely industry-based (classical) AI practitioners, sharing insights from her research and showcasing the research being done at CSIRO in QML. She also facilitated a panel discussion on Quantum Generative AI. Farina Riaz was one of the invited panelists to the discussion that focused on the potential of Quantum Generative AI, the relationship between classical and quantum AI, how AI professionals can get involved in quantum AI, and how to improve the gender balance in AI and quantum computing.

  • Mohan Baruwal Chhetri gave an invited talk at the FS-ISAC ANZ Member Forum on “Collaborative Intelligence in Security Operations Centres” on 30 May 2024.
  • Mohan Baruwal Chhetri gave an invited talk on Quantum Readiness of Australian Industries to the Australian Research Council Centre of Excellence for Engineered Quantum Systems (EQUS) Seminar Series on 2 July 2024. 
  • Mohan Baruwal Chhetri attended the Responsible Innovation Future Science Platform Showcase in Brisbane on 12 July 2024.

  • Shahroz Tariq gave a seminar on Human-AI Collaboration at Sungkyunkwan University, South Korea in May 2024.
  • Fatemeh Jalalvand gave a talk at the OPTIMA-CON 2024 in Melbourne titled “Enhancing Optimization Methods for Real- World Applications: Integrating Machine Learning Techniques for Improved Performance” (July 2024).
  • Raymond Zhao was an invited speaker for the Brainlink X-Lab Day 2024, South Korea, which ran from July 8-11, 2024, giving a talk on “High Throughput Lattice-based Signatures on GPUs: Comparing Falcon and Mitaka”. The goal of this conference was to expand the international network for the advancement and development of safe and reliable technologies, by inviting world-class scholars in the AI and cybersecurity field. The event was hosted by the Korean Federation of Science and Technology Societies (KOFST) and the Korean Ministry of Science ICT. The CQS Group’s Jason Xue was one of the organisers. More details are available here.
  • Shahroz Tariq, Sharif Abuadbba, Kristen Moore, and Simon S. Woo co-organised the 3rd Workshop on the security implications of Deepfakes and Cheapfakes (WDC ’24) co-located with ACM ASIACCS 2024 on July 2, 2024.

Regular Events