Research Intrerest
I believe deeply in the transformative power of collaboration. It has the potential to amplify human intelligence, as demonstrated in interdisciplinary research and cross-functional team innovation. Yet, effective collaboration is often hindered by barriers from differences in expertise, perspectives, motivations, and varying levels of engagement or trust, especially as scale and diversity increase. As teamwork evolves with AI as either part of the collaboration or the intervention, it brings new opportunities but also presents challenges—ensuring AI remains responsible and adaptive without compromising human ownership, creativity, and autonomy across different contexts, while truly enhancing human intelligence. These complexities drive my HCI journey and inspire my pursuit of a Ph.D. degree:
how can we enhance human intelligence by supporting 1) knowledge sharing, 2) collective decision-making, and 3) collective innovation?
My recent work
Exploring the Usage of Generative AI for Group Project-Based Offline Art Courses in Elementary Schools
Zhiqing Wang, Haoxiang Fan, Zhenhui Peng, submitted to CSCW 2025, under Major Revision
This study introduces a four-phase field study, involving in total two experienced K-6 art teachers and 132 students in eight offline course sessions, to investigate the usage and impact of GenAI in PBL(Project-Based Learning) for young leaners in real classroom setting. Our findings revealed the benefits of GenAI in providing background information, inspirations, and personalized guidance. However, challenges in query formulation for generating expected content were also observed. Moreover, students employed varied collaboration strategies, and teachers noted increased engagement alongside concern regarding misuse and interface suitability.
DesignWeaver: Dimensional scaffolding for text-to-image product design
Sirui Tao, Ivan Liang, Cindy Peng, Zhiqing Wang, Srishti Palani, Steven Dow, submitted to CHI 2025, under R&R
In this study, we developed DesignWeaver, an interface that helps novices generate prompts for a text-to-image model by surfacing key product design dimensions from generated images into a palette for quick selection. In a study with 52 novices, DesignWeaver enabled participants to craft longer prompts with more domain-specific vocabularies, resulting in more diverse, innovative product designs. However, the nuanced prompts heightened participants’ expectations beyond what current text-to-image models could deliver. We discuss implications for AI-based product design support tools.
CanAnswer: A community-powered conversational agent for learning about colorectal cancer
Yiwei Yuan, Zhiqing Wang, Zhenhui Peng, submitted to CHI 2025
This paper presents CanAnswer, a CA enhanced by community data for learning about colorectal cancer. We develop a computational workflow for processing community data to enrich the CA dataset with doctor-contributed documents and doctor-patient conversations, suggest follow-up questions, and provide related real-world cases. A between-subject study (N=24) shows that compared to the baseline condition with an LLM-based CA and raw community data, CanAnswer improves the recalled gained knowledge and reduces the task workload of the learning session. Our expert interviews (N=6) further confirm the reliability and usefulness of CanAnswer. We discuss the concerns and implications of using community data in health-related CAs.
Besides HCI
Social science
During my work on Chinese philanthropy with Prof. Edward Cunningham at Harvard, we conducted research into how donations from foundations, universities, and corporations reflect broader societal trends. Analyzing over 10,000 records, I uncovered shifts in donor behavior during the COVID-19 pandemic, such as an increase in multi-cause donations and a move from local giving to cross-provincial support.
Wireless communication
Advised by Prof. Danijela Cabric at UCLA, we conducted research on enhancing spectrum utilization in 5G networks using large-scale antenna technology and cognitive radio (CR). We designed a three-base-station system for full-space spectrum sharing, distinguishing CRs by angular information and reducing pilot contamination with a two-dimensional discrete Fourier Transform. A novel greedy CR scheduling algorithm demonstrated superior efficiency, successfully recognizing and scheduling over 100 CRs in less time compared to two-base-station setups.
Sustainable Energy
Advised by Prof. Feipeng Wang at Chongqing University, we conducted research on sustainable energy solutions, focusing on plant-based insulating fluids as eco-friendly alternatives to mineral oil in transformers. Our work reviewed the performance, challenges, and future potential of these biodegradable, non-toxic fluids, emphasizing their advantages in fire safety and environmental sustainability. This research highlights the critical role of plant-based oils in advancing safer and cleaner energy systems.
PUBLICATIONS
Tao, S., Liang, I., Peng, C., Wang, Z. , Palani, S., and Dow, S., 2025. DesignWeaver: Dimensional scaffolding for text-to-image product design. Submitted to the ACM Conference on Human Factors in Computing Systems (CHI’25, under R&R).
Wang, Z., Fan, H. and Peng, Z., 2025. Exploring the usage of generative AI for group project-based offline art courses in elementary schools. Submitted to the ACM Conference on Computer-Supported Cooperative Work and Social Computing (CSCW’25, under major revision).
Wang, J., Li, Z., Guo, Z., Wang, Z., Du, S., Gao, H., Du, W., Shi, D., He, L., Qi, L. and Chen, F., 2024. Enhancing decision credibility in transport safety through a modified IPSI–EXPROM Ⅱ–PAM model with kernel density estimation. Advanced Engineering Informatics, 62, p.102950.
Shen, Z., Wang, F., Wang, Z. and Li, J., 2021. A critical review of plant-based insulating fluids for transformer: 30-year development. Renewable and Sustainable Energy Reviews, 141, p.110783.