
Dr. Trung Le
Associate Professor · Monash University

Dr. Trung Le
Associate Professor
Faculty of Information Technology, Monash University, Australia
My research involves both theoretical and practical aspects. More specifically, this focuses on deep generative models, kernel methods, optimization in machine learning and Bayesian inference whose gained theories can be applied to supervised learning, semi-supervised learning, adversarial learning, online learning, anomaly detection, and cyber security. I have published in the top-notch conferences and high quality journals in machine learning, artificial intelligence, and data mining including NIPS, ICLR, AISTATS, UAI, IJCAI, ICDM and Journal of Machine Learning Research (JMLR).
News
Latest updates
Recent paper acceptances, preprints, and lab announcements.
Multiple papers accepted at ICML 2026 spanning anomaly detection, knowledge distillation, and preference optimization.
New preprint on safety alignment via density ratio matching (BSO) released.
Several works accepted at ACL 2026 on LLM distillation and multimodal robustness.
Paper on Multi-Cost Wasserstein Knowledge Distillation accepted at AAAI 2026.
Research
Research directions in artificial intelligence
From alignment and security to theory, reasoning, and efficient foundation models.
Human-AI Alignment
Preference optimization, safety alignment, and density-ratio methods for aligning foundation models with human intent.
ExploreExplainable & Efficient AI
Interpretable models and resource-efficient learning for deployable, trustworthy AI systems.
ExploreContinual Learning Foundation Models
Lifelong adaptation, rehearsal-free learning, and unlearning for evolving foundation models.
ExploreML/DL Theory
Kernel methods, optimization theory, and convergence analysis for modern deep learning.
ExploreFoundation Models Security
Jailbreak defense, adversarial robustness, and internal safeguards for LLMs and multimodal models.
ExploreDiffusion & Flow Matching
Generative modeling via diffusion, score-based methods, and flow matching for high-fidelity synthesis.
ExploreMachine Reasoning
Plan-guided learning, structured inference, and reasoning capabilities in large language models.
ExploreFoundation Distillation
Knowledge distillation, Wasserstein transfer, and efficient student models from large teachers.
ExploreContact
Get in touch
Dr. Trung Le, Associate Professor
Faculty of Information Technology, Monash University, Australia
I welcome enquiries regarding research collaboration. Please contact me by email.