About Me

I am a second-year PhD student in Information Engineering at the University of Cambridge, working in the Machine Intelligence Lab under the supervision of Prof. Bill Byrne. I am fully funded by the Cambridge International Scholarship from Cambridge Trust. Prior to this, I worked as a Machine Learning Engineer at To0 Space, a start-up developing diffusion-based models for high-fidelity architectural rendering.

I hold an M.Phil degree in Machine Learning and Machine Intelligence from the University of Cambridge, where I specialized in speech and language processing and supervised by Prof. Byrne.

I obtained my B.S degree from the University of California, San Diego with double majors in Computer Science and Cognitive Science and minor in Mathematics, where I was a member of the Caledonian Honors Society at Muir College and graduated Cum Laude.

My research interests include LLM/VLM Safety, LLM Alignment, Retrieval Augmented Generation, Vision-Language-Action (VLA) Models, Minimum Bayes Risk Decoding, and Neural Machine Translation.

Education

• Ph.D, Engineering, University of Cambridge (Oct 2024 - Present)
• M.Phil, Machine Learning and Machine Intelligence, University of Cambridge (Oct 2022-Aug 2023)
• B.S, Computer Science/Cognitive Science, UC San Diego (Oct 2018-Jun 2022), Cum Laude

Publications (Google Scholar)

1. Guangyu Yang, Jinghong Chen, Jingbiao Mei, Weizhe Lin, and Bill Byrne. Retrieval-Augmented Defense: Adaptive and Controllable Jailbreak Prevention for Large Language Models (ArXiv Preprint).
2. Jingbiao Mei, Jinghong Chen, Guangyu Yang, Weizhe Lin, and Bill Byrne. Robust Adaptation of Large Multimodal Models for Retrieval Augmented Hateful Meme Detection (EMNLP 2025 Oral).
3. Jinghong Chen, Guangyu Yang, Weizhe Lin, Jingbiao Mei, and Bill Byrne. On Extending Direct Preference Optimization to Accommodate Ties (NeurIPS 2025).
4. Guangyu Yang, Jinghong Chen, Weizhe Lin, and Bill Byrne. Direct Preference Optimization for Neural Machine Translation with Minimum Bayes Risk Decoding (NAACL 2024).
5. Guangyu Yang. Multilingual Models in Neural Machine Translation (MPhil Thesis).

Working Experience

Machine Learning Engineer @ To0 Space (Sep 2023 - Sep 2024)

• Fine-tuning Stable Diffusion (SD) model and Vision Transformer (ViT) for architecture rendering image.
• Retrieval Augmented Generation (RAG) for architecture rendering.
• Automatic labelling and captioning for architecture images with Vision Language Models (VLMs) and Large Language Models (LLMs).

Teaching Experience

Instructional Assistant @ UC San Diego

• COGS 118A. Supervised Machine Learning Algorithms