Ziang Xiao

Ziang Xiao is an Assistant Professor in the Department of Computer Science at Johns Hopkins University's Whiting School of Engineering and a member of the Data Science and AI Institute. His research bridges human-computer interaction, natural language processing, and social psychology to understand human behavior at scale through AI systems.

His educational background includes a Ph.D. in Computer Science from the University of Illinois Urbana-Champaign (co-advised by Prof. Hari Sundaram and Prof. Karrie Karahalios), where he also earned dual Bachelor's degrees in Psychology and Statistics & Computer Science under Prof. Dov Cohen. Prior to joining JHU, he was a postdoctoral researcher in the Fairness, Accountability, Transparency, and Ethics group at Microsoft Research Montréal.

Xiao's research focuses on three interconnected areas: AI for Social Science (using LLMs to simulate human behavior), Human-centered Model Evaluation (developing frameworks like ECBD for evidence-centered benchmark design), and Information Seeking (studying how AI systems affect information diversity). His work emphasizes democratizing technology to operationalize human intuitions about behavior and decision-making, with notable contributions to LLM evaluation metrics and ethical AI design.

Analysis of his recent publications reveals a strong trend toward human-centered AI evaluation, with increasing focus on LLM safety, value alignment, and interdisciplinary applications across social science domains. His work consistently appears in top-tier venues including CHI, ACL, and NeurIPS, with a 2024 CHI Best Paper Award for research on LLM-powered search systems.

• Best Paper Award at CHI 2024 for 'Generative Echo Chamber? Effect of LLM-Powered Search Systems on Diverse Information Seeking'
• Multiple publications in ACM/IEEE flagship conferences (CHI, ACL, NeurIPS)
• Research featured in Johns Hopkins news for social robot interruption handling and chatbot bias studies

Xiao actively advises doctoral students and postdocs, with current advisees including Yu Lu Liu, Nikhil Sharma, and Han Jiang. His teaching includes graduate courses on Human-Computer Interaction and Advanced HCI Research Methods. He co-leads research initiatives at the intersection of AI and social science, collaborating with the Data Science and AI Institute to develop frameworks for responsible AI deployment. Current projects examine LLM vulnerabilities in GUI agents, multilingual information disparities, and adaptive decision support systems that balance automation with user control.