Van Nguyen

I am Van Nguyen, a Postdoctoral Research Fellow in the Department of Software Systems & Cybersecurity at Monash University, Australia, under the supervision of Prof Carsten Rudolph. I am also an Affiliate at CSIRO’s Data61, Australia. I completed my PhD in Computer Science at Monash University in 2021, under the supervision of Prof Dinh Phung and Dr Trung Le. My thesis, “Deep Learning for Software Security”, can be accessed here. In the same year, I was recognized as a global talent through Australia’s Global Talent Independent Program (GTI), thanks to the recommendation of Prof John Grundy.

To date, I have 5+ years of experience in computer science, data science, cybersecurity, and software engineering. My work in these areas has resulted in 20+ publications in leading conferences and journals, including ESEC/FSE, ICLR, TSE, TOSEM, KDD, IJCNN, PAKDD, ASIACCS, and UAI.

My current research focuses on advancing machine learning, deep learning, and large language model-based approaches to solve topical problems in network and software security, including out-of-distribution code identification in software systems, robust and multilingual software vulnerability detection with explanations, transparent and reliable automated vulnerability repair, efficient and secure code generation, and adaptive AI solutions for emerging threats, aiming for safer and more reliable software development, quality assurance, and enhanced security systems. I am also working on addressing phishing attacks and deepfakes, with a particular focus on enhancing the robustness, scalability, and explainability of detection methods. In addition, I am advancing LLM-based solutions to the Text-to-SQL problem, which aims to enable natural language interfaces for querying databases, an area with significant industrial relevance and practical impact, a line of research aligned with my interests in using AI to solve real-world problems. My long-term research goal is to advance AI systems to perform complex human learning, explanation, and reasoning, enabling more intelligent, adaptive, and efficient solutions.

Alongside my expertise in theories and methodologies of computer science, data science, cybersecurity, and software engineering, I have solid programming and technical skills, with proficiency in Python and a variety of libraries and frameworks such as TensorFlow, Keras, PyTorch, Scikit-learn, NumPy, Pandas, Matplotlib, Seaborn, and Hugging Face for data analysis and implementing machine learning, deep learning, and large language models. I am also experienced in Power BI, SQL, C/C++, Git, R, Linux, and CUDA.

Selected News

[Dec 2025] I am happy to share that I have received the Postdoctoral Research Fellow Excellence 2025 Award from the Department of Software Systems and Cybersecurity (SSC), Faculty of Information Technology, Monash University, recognising my research excellence, impact, and service contributions.

[Oct 2025] I am pleased to share our latest research, “MulVuln: Enhancing Pre-trained LMs with Shared and Language-Specific Knowledge for Multilingual Vulnerability Detection”. In this study, we address the multilingual vulnerability detection problem, a challenging and relatively underexplored area in automated, AI-driven software security. The paper preprint is now available at https://arxiv.org/pdf/2510.04397.

[Oct 2025] I am honored to serve as an Area Chair for ICLR 2026 and a Program Committee Member for AAAI 2026, both top-tier A* conferences in machine learning.

[May 2025] I am thrilled to share that our paper, “SAFE: A Novel Approach For Software Vulnerability Detection from Enhancing The Capability of Large Language Models”, has been accepted at the 20th ACM ASIA Conference on Computer and Communications Security (ACM ASIACCS 2025), one of the top conferences in cybersecurity. In this work, we propose an innovative approach to enhance the capability of large language models in learning the semantic and syntactic relationships within source code, enabling more robust software vulnerability detection. I am deeply grateful to my collaborators for their invaluable feedback and support throughout this research.

[Mar 2025] I am excited to share that, with the support of Monash University, my proposal titled “Robust and Adaptive AI-driven Solutions for Software Development and Quality Assurance” has been submitted to the Australian Research Council (ARC) Discovery Early Career Researcher Award (DECRA) 2026. This project aims to create innovative solutions to address the limitations of current methods in software development and quality assurance practices. Ultimately, the solutions help create scalable, high-quality, secure, and adaptive software with minimal operational costs while preventing catastrophic system failures and financial losses caused by software vulnerabilities.

[Jan 2025] I am delighted to announce that our paper, “AI2TALE: An Innovative Information Theory-based Approach for Learning to Localize Phishing Attacks”, has been accepted at the International Conference on Learning Representations (ICLR 2025), a top-tier A* conference in machine learning. In this work, we propose an innovative deep learning-based method to solve the phishing attack localization problem, with the aim of improving the explainability (transparency) of email phishing detection. Our proposed method is lightweight, effective, and practical. Leading this work as the first and corresponding author has been an incredibly rewarding experience. I am grateful to my collaborators for their valuable feedback throughout this research.

[Dec 2024] For my role as Head Teaching Associate for the Deep Learning unit (with nearly 300 students enrolled) in the Department of Data Science and AI at Monash University, my teaching received a “very high” student satisfaction rating of 90/100, according to the Teaching Evaluation Report.

[Aug 2024] The research proposal, “Continual Learning of Large Language Models for Version-Controlled Code Generation”, in collaboration with my colleagues Dr Trang Vu and Dr Tongtong Wu, secure the Monash FIT Early Career Academics Seed Grant.

[May 2024] I had the privilege of serving as a guest lecturer for the Software Security unit in the Department of Software Systems and Cybersecurity at the Faculty of Information Technology, Monash University, where I presented current challenges and cutting-edge solutions in software security.

[Apr 2024] I was pleased to share that our paper, “Deep Domain Adaptation with Max-Margin Principle for Cross-Project Imbalanced Software Vulnerability Detection”, was accepted in ACM Transactions on Software Engineering and Methodology (TOSEM), a prestigious Q1-ranked journal in software engineering. In this paper, we propose an innovative solution that leverages solid theories of kernel methods and deep domain adaptation for cross-project imbalanced software vulnerability detection. It has been a truly rewarding experience to lead this work as the first and corresponding author. I sincerely appreciate the insightful feedback and support from my collaborators throughout this research.

Selected Publications

2025:

Van Nguyen, Surya Nepal, Xingliang Yuan, Tingmin Wu, Fengchao Chen, and Carsten Rudolph. MulVuln: Enhancing Pre-trained LMs with Shared and Language-Specific Knowledge for Multilingual Vulnerability Detection. https://arxiv.org/pdf/2510.04397.

Michael Fu, Trung Le, Van Nguyen, Chakkrit Tantithamthavorn, and Dinh Phung. Deepvulmatch: Learning and matching latent vulnerability representations for dual-granularity vulnerability detection. IEEE Transactions on Reliability, 2025. (Rank: Q1)

Van Nguyen, Surya Nepal, Xingliang Yuan, Tingmin Wu, and Carsten Rudolph. SAFE: A Novel Approach For Software Vulnerability Detection from Enhancing The Capability of Large Language Models. The ACM ASIA Conference on Computer and Communications Security (ACM ASIACCS), 2025. (Rank: A)

Van Nguyen, Tingmin Wu, Xingliang Yuan, Marthie Grobler, Surya Nepal, and Carsten Rudolph. AI2TALE: An Innovative Information Theory-based Approach for Learning to Localize Phishing Attacks. The International Conference on Learning Representations (ICLR), 2025. (Rank: A*)

Bowen Zhang, Hui Cui, Van Nguyen, and Monica Whitty. Audio Deepfake Detection: What Has Been Achieved and What Lies Ahead. Sensors, 2025. (Rank: Q1 (Instrumentation)).

2024:
Van Nguyen, Trung Le, Chakkrit Kla Tantithamthavorn, John Grundy, and Dinh Phung. Deep Domain Adaptation With Max-Margin Principle for Cross-Project Imbalanced Software Vulnerability Detection. ACM Transactions on Software Engineering and Methodology (TOSEM), 2024. (Rank: Q1)

Michael Fu, Van Nguyen, Chakkrit Kla Tantithamthavorn, Trung Le, and Dinh Phung. Vision transformer-inspired automated vulnerability repair. ACM Transactions on Software Engineering and Methodology (TOSEM), 2024. (Rank: Q1)

Michael Fu, Chakkrit Tantithamthavorn, Trung Le, Yuki Kume, Van Nguyen, Dinh Phung, and John Grundy. AIBugHunter: A Practical Tool for Predicting, Classifying and Repairing Software Vulnerabilities. Empirical Software Engineering, 2024. (Rank: Q1)

2023:
Michael Fu, Van Nguyen, Chakkrit Kla Tantithamthavorn, Trung Le, and Dinh Phung. VulExplainer: A Transformer-based Hierarchical Distillation for Explaining Vulnerability Types. IEEE Transactions on Software Engineering (TSE), 2023. (Rank: Q1)

Vy Vo, Van Nguyen, Trung Le, Quan Hung Tran, Reza Haf, Seyit Camtepe, and Dinh Phung. An Additive Instance-Wise Approach to Multi-class Model Interpretation. The International Conference on Learning Representations (ICLR), 2023. (Rank: A*)

Vy Vo, Trung Le, Van Nguyen, He Zhao, Edwin Bonilla, Gholamreza Haffari, and Dinh Phung. Feature-based Learning for Diverse and Privacy-Preserving Counterfactual Explanations. 29th ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD), 2023. (Rank: A*)

2022:
Van-Anh Nguyen, Dai Quoc Nguyen, Van Nguyen, Trung Le, Quan Hung Tran, and Dinh Phung. ReGVD: Revisiting Graph Neural Networks for Vulnerability Detection. IEEE/ACM 44th International Conference on Software Engineering: Companion Proceedings (ICSE-Companion), 2022.

Michael Fu, Chakkrit Tantithamthavorn, Trung Le, Van Nguyen, and Dinh Phung. VulRepair: A T5-Based Automated Software Vulnerability Repair. 30th ACM Joint European Software Engineering Conference and Symposium on the Foundations of Software Engineering (ESEC/FSE), 2022. (Rank: A*)

Tuan Nguyen, Van Nguyen, Trung Le, He Zhao, Quan Hung Tran, and Dinh Phung. Cycle Class Consistency with Distributional Optimal Transport and Knowledge Distillation for Unsupervised Domain Adaptation. The Conference on Uncertainty in Artificial Intelligence (UAI), 2022. (Rank: A*)

2021:
Van Nguyen, Trung Le, Olivier De Vel, Paul Montague, John Grundy, and Dinh Phung. Information-theoretic source Code Vulnerability Highlighting. International Joint Conference on Neural Networks (IJCNN), 2021.

2020:
Van Nguyen, Trung Le, Tue Le, Khanh Nguyen, Olivier de Vel, Paul Montague, and Dinh Phung. Code Pointer Network for Binary Function Scope Identification. International Joint Conference on Neural Networks (IJCNN), 2020. (Rank: A)

Van Nguyen, Trung Le, Olivier de Vel, Paul Montague, John Grundy, and Dinh Phung. Dual-Component Deep Domain Adaptation: A New Approach for Cross Project Software Vulnerability Detection. The Pacific-Asia Conference on Knowledge Discovery and Data Mining (PAKDD), 2020. (Rank: A)

Van Nguyen, Trung Le, Tue Le, Khanh Nguyen, Olivier de Vel, Paul Montague, John Grundy, and Dinh Phung. Code Action Network for Binary Function Scope Identification. Advances in Knowledge Discovery and Data Mining: 24th Pacific-Asia Conference, PAKDD, 2020. (Rank: A)

2019:
Van Nguyen, Trung Le, Tue Le, Khanh Nguyen, Olivier DeVel, Paul Montague, Lizhen Qu, and Dinh Phung. Deep Domain Adaptation for Vulnerable Code Function Identification. International Joint Conference on Neural Networks (IJCNN), 2019. (Rank: A)

Services

2025:
Area Chair: Serving as an area chair for ICLR (one of the top-tier (A*) conferences in machine learning).

2024–Present:
Program Committee Member: The Association for the Advancement of Artificial Intelligence (AAAI).

2021–Present:
Reviewer for leading conferences and journals in machine learning, cybersecurity, and software engineering, including:

Journals: Automated Software Engineering; Computers & Security; IEEE Transactions on Software Engineering (TSE); ACM Transactions on Software Engineering and Methodology (TOSEM).

Conferences: The Web Conference (WWW); International Conference on Learning Representations (ICLR); Asian Conference on Machine Learning (ACML); Neural Information Processing Systems (NeurIPS); Uncertainty in Artificial Intelligence (UAI); International Conference on Machine Learning (ICML).