Wenkang Jiang

About Me

I am a PhD student at the Australian Institute for Machine Learning (AIML), Adelaide University, where I am a member of the Causal AI Group. I am supervised by Prof. Javen Qinfeng Shi, Dr. Yuhang Liu, and Assoc. Prof. Ehsan Abbasnejad. Before commencing my PhD, I received an M.S. in Computer Technology (2024) from the University of Chinese Academy of Sciences, based at the Computer Network Information Center, Chinese Academy of Sciences, where my M.S. research was supervised by Sr. Eng. Hongbo He. I received a B.S. in Computer Science and Technology (2021) from Hefei University of Technology.

Research Interests

My research centers on causality, with a particular focus on causal representation learning as a principled approach to uncovering latent mechanisms and making black-box models more interpretable and reliable. I apply these ideas to single-cell biology, where I study the mechanisms governing cellular responses to interventions, and to LLM safety, where I investigate the internal representations and computational mechanisms underlying model behavior and failure modes. Ultimately, I aim to develop causally grounded methods for understanding, predicting, and steering complex learning systems.

• Causal Representation Learning: learning latent causal structures and mechanisms from observational and interventional data to enable more interpretable and generalizable learning.
• AI for Science: developing machine learning methods for biological discovery, with a focus on representation learning and cellular response prediction from single-cell perturbation data.
• LLM Safety and Interpretability: understanding and steering the internal representations and mechanisms underlying model behavior, vulnerabilities, and safety failures.