Will Wang

Huichen Will Wang

I'm a second-year Computer Science PhD student at the University of Washington. My work sits at the intersection of information visualization, human-computer interaction, and generative AI. I am fortunate to be co-advised by Jeffrey Heer and Leilani Battle.

Prior to UW, I did my undergrad at Amherst College. I had many great mentors, including Cindy Bearfield, Kate Follette, and Lee Spector.

Email / Google Scholar / Twitter / LinkedIn

Research

My research focuses on developing and evaluating human-centric tools for information visualization and data science.

Chef's Choice

* denotes equal contribution.

	FullFront: Benchmarking MLLMs Across the Full Front-End Engineering Workflow Haoyu Sun, Huichen Will Wang, Jiawei Gu, Linjie Li, Yu Cheng Under Review, 2025 PDF / arXiv / code We contribute FullFront, a benchmark spanning the full front-end development pipeline: Webpage Design, Webpage Perception, and Webpage Code Generation.
	Do You "Trust" This Visualization? An Inventory to Measure Trust in Visualizations Huichen Will Wang, Kylie Lin, Andrew Cohen, Ryan Kennedy, Zach Zwald, Carolina Nobre, Cindy Xiong Bearfield Under Review, 2025 PDF / arXiv Through Exploratory Factor Analysis, we derive an operational definition of trust in visualizations and contribute an inventory to measure it.
	Grid Labeling: Crowdsourcing Task-Specific Importance from Visualizations Minsuk Chang, Yao Wang, Huichen Will Wang, Andreas Bulling, Cindy Xiong Bearfield Eurographics Conference on Visualization (EuroVis Short Paper), 2025 PDF / arXiv / code We introduce Grid Labeling—a novel annotation method for collecting task-specific "saliency" on visualizations.
	Can MLLMs Reason in Multimodality? EMMA: An Enhanced MultiModal ReAsoning Benchmark Yunzhuo Hao, Jiawei Gu, Huichen Will Wang, Linjie Li, Zhengyuan Yang, Lijuan Wang, Yu Cheng International Conference on Machine Learning (ICML), 2025 (Oral Presentation, top 0.9%) Project Page / PDF / arXiv We contribute EMMA (Enhanced MultiModal reAsoning), a benchmark targeting organic multimodal reasoning across mathematics, physics, chemistry, and coding. SOTA models struggle big time.
	Jupybara: Operationalizing a Design Space for Actionable Data Analysis and Storytelling with LLMs Huichen Will Wang, Larry Birnbaum, Vidya Setlur Conference on Human Factors in Computing Systems (CHI), 2025 PDF / code We synthesize a design space for actionable exploratory data analysis and storytelling, operationalizing it through Jupybara—a Jupyter Notebook plugin featuring a multi-agent system. PSA: Jupybara = Jupyter Notebook + Capybara
	How Aligned are Human Chart Takeaways and LLM Predictions? A Case Study on Bar Charts with Varying Layouts Huichen Will Wang, Jane Hoffswell, Sao Myat Thazin Thane, Victor S. Bursztyn, Cindy Xiong Bearfield IEEE Transactions on Visualization and Computer Graphics (Proceedings of VIS), 2025 PDF / arXiv Human chart takeaways are sensitive to design choices in a visualization. LLMs struggle to replicate this sensitivity, often generating takeaways that don't match human interpretation patterns.
	DracoGPT: Extracting Visualization Design Preferences from Large Language Models Huichen Will Wang, Mitchell Gordon, Leilani Battle, Jeffrey Heer IEEE Transactions on Visualization and Computer Graphics (Proceedings of VIS), 2025 PDF / arXiv / code We contribute DracoGPT, a method for extracting, modeling, and assessing visualization design preferences from LLMs. LLMs' design preferences diverge from guidelines drawn from human subjects experiments.

Website design adapted from Jon Barron.