Zhiyu Wang | 王智语

👋 Hello! I’m thrilled you’re here. Here’s a snapshot of my journey so far:

Current

I am currently working as a Research Assistant at Shanghai Jiao Tong University focusing on protein local structure alignment.

Background

At University of Cambridge, I completed my M.Phil. in Advanced Computer Science with Distinction (ranked 8th out of 60). My dissertation Topotein: Topological Deep Learning for Protein Representation Learning, supervised by Prof. Pietro Liò, received a score of 88/100 and won the Highly Commended M.Phil Project 2024-2025 prize.

I previously graduated with First-Class Honors in B.Sc. Computer Science from University College London, where I completed my dissertation on ML-based remote patient monitoring under Prof. Ivana Drobnjak.

In industry, I worked as a Machine Learning Engineering intern at Xunfei Healthcare, focusing on LLM instruction-tuning. I also collaborated with IBM and the NHS to develop chatbots for healthcare facilities.

Research Interests

I am interested in geometric deep learning and large language models. I’m dedicated to creating novel ML methods that address pressing biomedical and healthcare challenges. I welcome opportunities to collaborate on innovative, high-impact projects!

News

Jun 19, 2025	I am delighted to announce that I have been awarded a £350 Churchill College Postgraduate Academic Travel Fund (PAT0062) to attend IJCAI25 and present my GraphAU-Pain paper.
Jun 13, 2025	Our GraphAU-Pain paper has been accepted for presentation at the MiGA workshop at IJCAI25. Looking forward to seeing everyone in Guangzhou, China on August 28th!

Selected Publications

IJCAI MiGA
GraphAU-Pain: Graph-based Action Unit Representation for Pain Intensity Estimation

Zhiyu Wang, Yang Liu, and Hatice Gunes

In IJCAI Micro-gesture Analysis for Hidden Emotion Understanding (MiGA) Workshop, 2025

Abs arXiv Bib

Understanding pain-related facial behaviors is essential for digital healthcare in terms of effective monitoring, assisted diagnostics, and treatment planning, particularly for patients unable to communicate verbally. Existing data-driven methods of detecting pain from facial expressions are limited due to interpretability and severity quantification. To this end, we propose GraphAU-Pain, leveraging a graph-based framework to model facial Action Units (AUs) and their interrelationships for pain intensity estimation. AUs are represented as graph nodes, with co-occurrence relationships as edges, enabling a more expressive depiction of pain-related facial behaviors. By utilizing a relational graph neural network, our framework offers improved interpretability and significant performance gains. Experiments conducted on the publicly available UNBC dataset demonstrate the effectiveness of the GraphAU-Pain, achieving an F1-score of 66.21% and accuracy of 87.61% in pain intensity estimation.
@inproceedings{wang2025graphaupaingraphbasedactionunit, title = {GraphAU-Pain: Graph-based Action Unit Representation for Pain Intensity Estimation}, author = {Wang, Zhiyu and Liu, Yang and Gunes, Hatice}, year = {2025}, booktitle = {IJCAI Micro-gesture Analysis for Hidden Emotion Understanding (MiGA) Workshop}, address = {Guangzhou, China}, eprint = {2505.19802}, archiveprefix = {arXiv}, primaryclass = {cs.LG}, }