Ziwei Nina Chen

My name is Ziwei Nina Chen. I am a second-year PhD student in Computer Science & Engineering at UC San Diego, advised by Prof. Kristen Vaccaro. I am also a member of the Design Lab. I study how AI systems behave in human-centered contexts and how we can evaluate and design them to better support people. I am interested in developing evaluation methods and interactive systems that help us understand AI behavior and improve its alignment with human needs and social contexts.

Email CV Scholar Github

News

Sept 2026

I recently completed my fellowship at Eigen Labs, where I developed a benchmark for evaluating how speech-to-speech models understand non-speech vocalizations (e.g. laughs, sighs). It is a step toward more socially aware voice agents!

More news

Publications & Reports

	NSV-Shift: A Contrastive Benchmark for Non-Speech Vocalization Understanding and Response Adaptation in Speech-to-Speech Models PDF Ziwei Chen Eigen Labs Fellowship Report Non-speech vocalizations (NSVs), such as laughter, sighs, and gasps, carry emotional meaning in spoken conversation, yet benchmarks for whether speech-to-speech (S2S) models can interpret them and adapt their responses remain limited. We developed NSV-Shift, a contrastive benchmark framework for evaluating NSV understanding and response adaptation in S2S models. We built an end-to-end data construction and evaluation pipeline, validated it on an initial pilot set of 22 contrastive pairs, and conducted preliminary evaluations across five S2S models.
	Deception at Scale: Deceptive Designs in 1K LLM-Generated E-Commerce Components PDF Ziwei Chen, Jiawen Shen, Luna, Hanyu Zhang, Kristen Vaccaro CHI 2026 Recent work has shown that front-end code generated by Large Language Models (LLMs) can embed deceptive designs. To assess the magnitude of this problem, identify the factors that influence deceptive design production, and test strategies for reducing deceptive designs, we carried out two studies which generated and analyzed 1,296 LLM-generated web components, along with a design rationale for each.
	Where’s the Water? Supporting Clean Water Access for the Homeless Community PDF Video Presentation Winner Posts Alexandra Balmaceda, Ziwei Chen (* denotes equal contribution) CHI EA 2024 \| CHI Student Design Competition Winner (top 2 out of 43 submissions) Access to clean water is essential, yet it poses a significant challenge for the homeless population. Our project, "Where's the Water" is a web-based tool designed to improve water access for the homeless community. It maps nearby clean water sources like drinking fountains, public restrooms, and showers. The tool's design was informed through interviews with the homeless community in Ann Arbor, Michigan.