Hi! πŸ‘‹ My name is Jiayi Ye (叢佳一).

I am a final-year undergraduate student at Sichuan University, actively seeking a PhD position for the Fall 2026 intake. πŸŽ“

My research lies at the intersection of Natural Language Processing (NLP) and Human-Computer Interaction (HCI), with a particular focus on applying technology to address broad socio-psychological needs. Globalization and social atomization have intensified widespread feelings of loneliness and social isolation, giving rise to a significant demand for emotional companionship. In response, my work is dedicated to designing and developing next-generation conversational agents to provide emotional support and accessible daily companionship for a broader population.

In this direction, I am especially interested in AI-driven virtual live streaming (AI VTuber) as a scalable intervention. As a highly interactive medium, live streaming holds the potential to deliver companionship at scale. It can serve a large number of users simultaneously through a one-to-many format, reducing individual social pressure while gathering like-minded viewers into an online community with a sense of belonging. I conducted a user study with viewers of AI VTubers to understand how they discover, bond with, and co-create around these virtual streamers. See our paper: My Favorite Streamer is an LLM (CHI 2026). I am also a maintainer for the Open-LLM-VTuber project (4.5k+ ⭐), a popular open-source tool for creating LLM-driven virtual streamers.

Previously, I was a research intern at the MINE Lab at the University of Notre Dame, advised by Prof. Xiangliang Zhang. I worked on Trustworthy AI and had a wonderful time there, contributing to projects like LLM Judge Bias (ICLR 2025) and TrustEval (NAACL 2025 Demo).

Fun fact: I'm also an AI VTuber fan myself! My favorite is Neuro-sama β€” and yes, my avatar is her. πŸ’œ