Yiran (Jenny) Shen

Hi, welcome to my homepage! I’m a second-year PhD student in Computer Science at the University of California, San Diego, advised by Prof. Prithviraj (Raj) Ammanabrolu at Pearls Lab🧋.

My research focuses on trustworthy AI for language models and agents, with an emphasis on pluralistic alignment, reasoning, and interpretability. I develop methods that help systems infer user intent through interaction, formalize it into clear, scalable evaluation criteria, and use these signals to train and assess agents on complex, long horizon tasks where success is difficult to verify. I also study reinforcement learning for adaptive multi-agent systems that support effective human-AI collaboration while respecting diverse values and constraints. My long term goal is to build safe, controllable agentic systems that reliably accomplish economically valuable work, and to understand their social and ethical implications in real decision-making.

I have previously interned at Alibaba Tongyi Lab, where I was advised by Dr. Liuyi Yao, working on research related to personalized alignment, role-play agents, and reasoning. Prior to joining UCSD, I completed my Master’s in Interdisciplinary Data Science at 💙 Duke University, where I was fortunate to work with Prof. Brandon T. Fain and Prof. Bruce W. Rogers. Before that, I received my Bachelor’s degree from the 🇦🇺 University of Melbourne, majoring in Economics and Finance with a minor in Statistics.