I am a Ph.D. candidate in Engineering at the University of Sydney, specializing in multimodal learning. I hold a Bachelor of Advanced Computing (Honours) in Computer Science and Computational Data Science from the University of Sydney (GPA 3.91/4.00, First-Class Honours, top 1%) and am supported by an Australian Government RTP Scholarship. During my undergraduate studies, I received the Ian Jackson Memorial Prize, was named a Dalyell Scholar, and consistently appeared on the Dean’s List.
My research on multimodal learning mainly focuses on:
(1) Multimodal collaborative decision-making (e.g., PET-CT with electronic health record);
(2) Cross-modality assisted learning (e.g., language-guided vision tasks);
(3) Multimodal generative methods for diagnostic support (e.g., medical visual question answering, radiology report generation).
🔥 News
- 2025.10: A paper on medical report generation is accepted to TMM
- 2025.03: A paper on textual reliance is accepted to ICCV 2025
- 2024.10: A paper on modality preference bias is accepted to the workshop in ACM MM 2024
- 2024.05: A paper on medical language-guided segmentation is accepted to MICCAI 2024
- 2024.03: I commenced my PhD in Engineering at the University of Sydney, focusing on multimodal learning and funded by an Australian Government RTP Scholarship.
- 2023.12: I graduated from the University of Sydney with a Bachelor of Advanced Computing (First Class Honours; WAM 95+)
- 2021.01: I was invited to be a Dalyell Scholar.
- 2022.09: I joined The University of Sydney as a research assistant.
- 2022.02: I participate in ICPC on behalf of The University of Sydney
📝 Publications
Dynamic Traceback Learning for Medical Report Generation
Shuchang Ye, Mingyuan Meng, Mingjian Li, Dagan Feng, Usman Naseem, Jinman Kim
Shuchang Ye, Usman Naseem, Mingyuan Meng, Jinman Kim
A Causal Approach to Mitigate Modality Preference Bias in Medical Visual Question Answering
Shuchang Ye, Usman Naseem, Mingyuan Meng, Dagan Feng, Jinman Kim
Enabling Text-Free Inference in Language-Guided Segmentation of Chest X-Rays via Self-guidance
Shuchang Ye, Mingyuan Meng, Mingjian Li, Dagan Feng, Jinman Kim
🎖 Honors and Awards
- 2024.03 Australia Government RTP Scholarship.
- 2023.09 Dean’s List excellence in academic performance
- 2023.07 The Ian Jackson Memorial Prize for Computer Science
- 2021.01 Dalyell Scholar
📖 Educations
- 2024.03 - Present, Doctor of Philosophy (Engineering), The University of Sydney
- Thesis: Multimodal Learning
- 2020.03 – 2024.03: Bachelor of Advanced Computing (Honours), The University of Sydney
- Major: Computer Science, Computational Data Science
💻 Work Experience
- 2026.01 - Present, Machine Learning Engineer, TikTok (ByteDance), Sydney, Australia
- 0->1 development of TikTok’s first Audio-Visual-Language foundation model (AVLM) for video and live moderation: data pipeline, pre-training, post-training, and downstream supervised fine-tuning and alignment.
- Deployed to production moderation systems, significantly reducing Community Guidelines Violation Rate (CGVR) and Creator Overkill Rate (COR) compared to existing ASR+VLM approach.
- 2022.07 - 2022.12, Research Assistant, The University of Sydney, Sydney, Australia
- Built the first facial recognition system for animals, demonstrating the potential to replace traditional and widely used tagging methods. PDF
- Key areas: computer vision, object detection, convolutional neural networks, and facial recognition.
📜 Reviewer Certificates
- 2025, IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI), Certificate