AMS Special Seminar Series | Xinran Wang

Location: Shaffer 3

When: February 19th at 1:30 p.m.

Title: Building Trust for Human-AI Coworking

Abstract: As AI systems transition from automation tools to autonomous collaborators, building user trust remains a fundamental challenge: users face repeated misalignment with their preferences, unreliable outputs mixed with hallucinations, and sycophantic responses lacking critical thinking. This talk presents a research agenda for building trustworthy human-AI coworking systems, along with a deep dive into multi-dimensional alignment of large language models.

Bio: Xinran Wang is a Ph.D. candidate in Computer Science at the University of Minnesota. Her research focuses on trustworthy AI systems for human–AI coworking. She is supported by the 3M Science and Technology Graduate Fellowship and has received the University of Minnesota Doctoral Dissertation Fellowship, the CS-IDEA Fellowship, and the John N. Quiring Fellowship in Statistics. She has published as a first author at top AI venues including ICLR (Oral, Spotlight), NeurIPS (Spotlight), and NAACL, and received the IEEE BigData Best Student Paper Award. Prior to her Ph.D., she worked in finance, technology, and consumer goods industries. She holds an M.S. in Statistics from the University of Minnesota and a Bachelor’s degree in Economics from Tsinghua University.

Zoom link: https://wse.zoom.us/j/92215066845