I am a PhD candidate in Computing and Information Systems at the University of Melbourne,
supervised by Prof. Christopher Leckie,
Prof. Sarah M. Erfani, and
Dr. Hanxun Huang.
My research focuses on trustworthy AI, with an emphasis on robustness, fairness,
adversarial reliability, and hallucination detection in vision-language models.
I am particularly interested in understanding how multimodal models fail under
distribution shifts, spurious correlations, adversarial perturbations, and visually
unsupported generation, and in developing practical methods to improve their reliability.
News
August 2026 "VisER: Visual Evidence and Reliance for Object Hallucination Detection in LVLMs" accepted at EMNLP 2026.
July 2026 Selected as one of 16 doctoral students for the Optiver PhD Quant Lab 2026.
June 2026 “GeoDetect: Geometric Adversarial Detection for VLPs” accepted at ECCV 2026.
June 2026 Awarded Women in Machine Learning travel funding and an FEIT travel scholarship.
May 2026 “Density-Aware Translation of Spurious Correlations in Zero-Shot VLMs” accepted at ICML 2026.