I am a third-year Ph.D. student in Computer Science at the Australian Institute for Machine Learning (AIML), Adelaide University (formerly The University of Adelaide), advised by Prof. Javen Qinfeng Shi. Before my Ph.D., I received my M.Sc. and B.Eng. degrees from Wuhan University of Technology.
I study how language supervision shapes the semantics, geometry, and identifiability of multimodal representations. My long-term goal is to develop principled foundations for language-guided representation learning, with the aim of building multimodal systems that acquire structured, interpretable, and reliable representations of the world.
ICML
International Conference on Machine Learning (ICML), 2026
A theoretical study of representation geometry in unimodal and multimodal contrastive learning, revealing a geometric bifurcation and identifying object-level causes of the modality gap.
ICLR
International Conference on Learning Representations (ICLR), 2026
An investigation of whether next-token prediction alone is sufficient for learning human-interpretable concepts from data.
NeurIPS
Advances in Neural Information Processing Systems (NeurIPS), 2025
Studies when controlled cross-modal misalignment can improve multimodal representation learning instead of only harming it.
ECCV
European Conference on Computer Vision (ECCV), 2024
Explores language-guided disentanglement of style and content through contrastive learning with augmented prompts.
Last updated: May 2026. · Inspired by Minimal Light.
