Longju Bai's Homepage

Hi! I’m a first-year PhD student in the Computer Science and Engineering Department at the University of Michigan, Ann Arbor, where I’m advised by Prof. Rada Mihalcea. I’m also fortunate to work closely with Prof. Jiaxin Pei (University of Texas, Austin) and Prof. Veronica Perez-Rosas (Texas State University). I’m deeply grateful for their invaluable insights and guidance.

Previously, I received my master’s degree in Electrical and Computer Engineering from the University of Michigan, Ann Arbor, specializing in Signal and Image Processing with a Machine Learning track.

My research interests lie broadly in Natural Language Processing and Multimodal AI, with the goal of building Transparent, Efficient, and Responsible intelligent systems. My current projects center on the following directions:

1. NLP for Mental Health and Counselor Training: AI holds significant promise for supporting mental health, yet its real-world impact remains underexplored. I’m investigating both direct AI-based support and its use in counselor training, with an emphasis on understanding real-world impact.

2. Coding Agents: Coding agents have advanced rapidly in the past year, but their token and resource costs are often opaque. I’m working on improving agent efficiency to reduce unnecessary cost, and on developing budget controllability so users can reliably constrain resource usage.

3. Transfer Learning and Domain Adaptation: AI systems often struggle on domains with limited data or distributional shifts, and single-stage fixes rarely generalize. I’m investigating methods spanning the full AI pipeline, including data synthesis, model training, and inference-time adaptation, to improve performance on these underrepresented domains.

Beyond these directions, I’m always excited to explore other topics that contribute to building more reliable AI systems. Please feel free to reach out if you’d like to discuss research or potential collaborations.