Kuei-Chun Kao

Hi! I am Johnson (Kuei-Chun) Kao, a PhD student studying Computer Science at University of California, Los Angeles (UCLA), advised by Prof. Cho-Jui Hsieh.

I am intern at Bytedance as Research Scientist Intern in Seattle.

My research interest lies in the intersection of Human Computer Interaction (HCI), Computer Vision (CV), and Natural Language Processing (NLP), aiming to equip computers with the ability to understand data across different modalities. Besides, I love to build efficient, reliable and scalable machine learning systems in different scenarios. Previously, I was fortune to work as Machine Learning Scientist with Prof. Min Sun at Appier. I received my B.Sc. degree from National Yang Ming Chiao Tung University where I was so lucky to be advised by Prof. Yung-Ju Chang, Prof. Wei-Chen Chiu, and Prof. Ching-Chun Huang. I was a Research Assistant at Institute of Information Science Academia Sinica, advised by Prof. Keh-Yih Su.

Specifically, I am interested in the following topcis:

Understanding and Benchmarking the capabilities and limitations of LLM/MLLMs: Instruction finetuning and Prompt optimization of (M)LLMs, Multimodal language agents, Commonsense Reasoning, T2I Reward Hacking
Human-AI Collaboration: Explainable AI, Ubiquitous computing, Developing robust and practical AI systems
Data Efficiency: Leveraging the generation ability of LLMs to create high-quality synthetic data to improve embodied agents and multimodal language models

Recent News

Nov, 2025: Our paper about T2I reward hacking is under review in CVPR 2025.
Oct, 2025: Our paper about Biomedical Auto Prompt Optimization is under review in Nature Intelligence.
Aug, 2025: Our paper QG-CoC is accepted in EMNLP 2025.
April, 2025: One paper got accepted in IJHCS 2025.
Sept, 2024: Our paper BeyondX is accepted in EMNLP 2024.
June, 2024: Publish an Arxiv Paper about LLM reasoing, please check out link.
Sep. 2023: Start Master at UCLA.
Jan. 2023: One paper got accpeted in CHI 2023.