Ziyi ‘Zoe’ Wang 王子一

Howdy! I’m Ziyi 👋 I’m a first-year Ph.D. student in Computer Science at Texas A&M University, advised by Prof. Meng Xia in the Dream Lab.

My research studies how well AI can model people, from simulating personas to inferring latent states such as engagement and affect, and how such models can help people understand others and make better decisions. I probe what models actually understand about people (VLMs & Engagement), build persona-based simulations that make diverse needs and trade-offs visible to decision-makers (StreetDesignAI), and design interactive systems that help people reflect and decide with AI (CareerPooler). My work appears at EMNLP, DIS, IEEE VIS, and AAAI, including a Best Paper at IEEE VIS 2026 and an oral presentation at EMNLP 2026.

Previously, I was a Master’s student in HCI at the University of Maryland, College Park, working with Dr. Zijian Ding and Prof. Fumeng Yang. I also collaborated with Prof. Yue Zhao, Prof. Xiyang Hu, and Prof. Xiang Yan.

Google Scholar Email LinkedIn Twitter

News

Sep 2026 Our IEEE VIS 2026 paper “Made to Feel” received the Best Paper Award! 🏆 Best Paper · Top 1%
Sep 2026 Our EMNLP 2026 paper “Do Vision Language Models Understand Human Engagement in Games?” was selected for an oral presentation. 🎤 Oral Presentation · Top 2.6%
Aug 2026 Started my Ph.D. in Computer Science at Texas A&M University, advised by Prof. Meng Xia. Howdy, College Station! 🎉 Milestone
Aug 2026 Our paper “Do Vision Language Models Understand Human Engagement in Games?” has been accepted to EMNLP 2026! See our preprint. Paper
Aug 2026 Our new paper “Physics-Preserving Latent Compression” has been accepted to ICDM 2026! It’s a physics-preserving latent compressor for 3D turbulence with zero-shot resolution transfer. See our preprint. Paper
Jun 2026 Presented StreetDesignAI at DIS 2026 in Singapore. Conference
Jun 2026 Our new paper “Made to Feel” has been accepted to IEEE VIS 2026! It’s an interview study of how designers bring emotions into affective visualization. See our preprint. Paper
May 2026 Graduated from the University of Maryland, College Park, with an M.S. in Human-Computer Interaction. Thank you, HCIL! 🎓 Milestone
May 2026 Our survey paper on a unified view of learning PDEs has been accepted to IJCAI 2026! See our preprint. Paper
Apr 2026 Attended CHI 2026 in Barcelona, Spain. Conference
Apr 2026 Our paper on multimodal ranking attacks against vision-language-model-based product search has been accepted to the ACL 2026 KnowFM Workshop! See our preprint. Paper
Mar 2026 We have a new paper examining whether vision-language models can infer human engagement from gameplay video. See our preprint. Preprint
Mar 2026 Our new paper “StreetDesignAI” has been accepted to DIS 2026! It’s a persona-based multi-agent system that surfaces experiential conflicts in cycling infrastructure design and supports inclusive, trade-off-aware decision making. See our preprint. Paper
Jan 2026 We have a new survey systematically reviewing 1,493 papers on human–AI decision-making, proposing unified taxonomies of AI and human roles. See our preprint. Preprint
Jan 2026 We have a new paper “Persona-aware and Explainable Bikeability Assessment” on persona-conditioned vision-language models for explainable bikeability prediction. See our preprint. Preprint
Jan 2026 We have a new paper introducing ResearchCube, a multi-dimensional trade-off exploration system for research ideation. See our preprint. Preprint
Nov 2025 Our paper on mitigating hallucinations in LLMs using causal reasoning has been accepted to AAAI 2026! See our preprint. Paper
Oct 2025 Our new paper “From Image Generation to Infrastructure Design” has been accepted to the NeurIPS 2025 UrbanAI Workshop! It’s a multi-agent pipeline for realistic street-design generation. See our preprint. Paper
Sep 2025 We have a new paper introducing CareerPooler, a generative-AI-powered pool-table metaphor system for career exploration that improves engagement, satisfaction, and clarity compared to chatbot baselines. See our preprint. Preprint
Jul 2025 Our new paper “Frontend Diffusion” has been accepted to IEEE VL/HCC 2025! It’s a multi-stage AI system that turns sketches into website code for junior researchers and designers. See our preprint. Paper
Jul 2025 Our new paper “JailDAM” has been accepted to COLM 2025! It proposes an adaptive memory approach for jailbreak detection in vision-language models. See the preprint. Paper
Jun 2025 We have a new paper accepted to ECML PKDD 2025 on leveraging LLMs for few-shot graph OOD detection. See our preprint. Paper
May 2025 We have a new paper on zero-shot graph OOD detection using foundation models (GLIP-OOD). See our preprint. Preprint
May 2025 We have a new paper introducing GOE-LLM, a framework using LLMs to generate synthetic OOD nodes for graph OOD detection without requiring real OOD data. See our preprint. Preprint

Publications

Full record on Google Scholar ↗

How well can AI model people, and how can such models help people understand others and make better decisions?

Understanding people Measuring and improving how LLMs and VLMs infer latent human states and perspectives: engagement from gameplay video, persona-conditioned judgments of the built environment, and the gap between perceiving visible cues and actually understanding people. VLMs & Engagement · Bikeability
Simulating perspectives Persona-based multi-agent simulation and affective visualization that make diverse needs, emotions, and trade-offs visible to the people making decisions, moving them from a single perspective toward deliberate trade-off reasoning. StreetDesignAI · Made to Feel
Working with AI How people reflect, create, and decide with AI, and when they should trust it: generative systems that scaffold career exploration, self-representation, and research ideation, a systematic review of human–AI decision-making, and methods that make model behavior more reliable and safe. CareerPooler · Frontend Diffusion · ResearchCube · Human–AI decision-making review · Causal reasoning for LLMs · JailDAM

EMNLP 2026 (Conference on Empirical Methods in Natural Language Processing)Oral Presentation · Top 2.6%

Do Vision Language Models Understand Human Engagement in Games?

Ziyi Wang^*, Qizan Guo^*, Rishitosh Singh^*, Xiyang Hu

PDF arXiv
DIS 2026 (ACM Designing Interactive Systems Conference)

StreetDesignAI: Broadening Designer Perspectives Through Multi-Persona Evaluation of Cycling Infrastructure

Ziyi Wang, Yilong Dai, Duanya Lyu, Mateo Nader, Sihan Chen, Wanghao Ye, Zijian Ding, Xiang Yan

PDF arXiv Video
arXiv 2025 (arXiv preprint)

CareerPooler: AI-Powered Metaphorical Pool Simulation Improves Experience and Outcomes in Career Exploration

Ziyi Wang, Ziwen Zeng, Yuan Li, Zijian Ding

PDF arXiv Video
arXiv 2026 (arXiv preprint)

Persona-aware and Explainable Bikeability Assessment: A Vision-Language Model Approach

Yilong Dai, Ziyi Wang, Chenguang Wang, Kexin Zhou, Yiheng Qian, Susu Xu, Xiang Yan

PDF arXiv
IEEE VIS 2026 (IEEE Visualization Conference (Short Paper))Best Paper · Top 1%

Made to Feel: How Designers Bring Emotions into Affective Visualization

Yixin Bai, Ziyi Wang, Keke Wu, Fumeng Yang

PDF arXiv
ICDM 2026 (IEEE International Conference on Data Mining)

Physics-Preserving Latent Compression for Zero-Shot Resolution Transfer in 3D Turbulence

Yilong Dai, Yiming Sun, Yiheng Chen, Ziyi Wang, Shengyu Chen, Xiaowei Jia, Runlong Yu

PDF arXiv
IJCAI 2026 (International Joint Conference on Artificial Intelligence (Survey Track))

Learning PDE Solvers with Physics and Data: A Unifying View of Physics-Informed Neural Networks and Neural Operators

Yilong Dai, Shengyu Chen, Ziyi Wang, Xiaowei Jia, Yiqun Xie, Vipin Kumar, Runlong Yu

PDF arXiv
arXiv 2026 (arXiv preprint)

ResearchCube: Multi-Dimensional Trade-off Exploration for Research Ideation

Zijian Ding, Fenghai Li, Ziyi Wang, Joel Chan

PDF arXiv
arXiv 2026 (arXiv preprint (demo paper))GitHub · 95 stars

Sovereign-OS: A Charter-Governed Operating System for Autonomous AI Agents with Verifiable Fiscal Discipline

Aojie Yuan, Haiyue Zhang, Ziyi Wang, Yue Zhao

PDF arXiv Code Demo
ACL 2026 KnowFM Workshop (Workshop on Towards Knowledgeable Foundation Models at ACL 2026)

Multimodal Generative Engine Optimization: Rank Manipulation for Vision-Language Model Rankers

Yixuan Du, Chenxiao Yu, Haoyan Xu, Ziyi Wang, Yue Zhao, Xiyang Hu

PDF arXiv
Preprint 2026 (HAL preprint)

In the Shadow of Judgment: Mapping Out the Landscape of Human–AI Decision-Making Through a Systematic Review

Yixin Bai, Taehyun Yang, Zhongzheng Xu, Dayeon Ki, Ziyi Wang, Yu Hou, Fumeng Yang

PDF HAL
ECML PKDD 2025 (European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases)

Few-Shot Graph Out-of-Distribution Detection with LLMs

Haoyan Xu, Zhengtao Yao, Yushun Dong, Ziyi Wang, Ryan A. Rossi, Mengyuan Li, Yue Zhao

PDF arXiv
IEEE VL/HCC 2025 (IEEE Symposium on Visual Languages and Human-Centric Computing)

Frontend Diffusion: Empowering Self-Representation of Junior Researchers and Designers Through Multi-agent System

Zijian Ding, Qinshi Zhang, Mohan Chi, Ziyi Wang

PDF arXiv Code Video
AAAI 2026 (AAAI Conference on Artificial Intelligence)

Mitigating Hallucinations in Large Language Models via Causal Reasoning

Yuangang Li, Yiqing Shen, Yi Nian, Jiechao Gao, Ziyi Wang, Chenxiao Yu, Shawn Li, Jie Wang, Xiyang Hu, Yue Zhao

PDF arXiv
COLM 2025 (Conference on Language Modeling)

JailDAM: Jailbreak Detection with Adaptive Memory for Vision-Language Model

Yi Nian, Shenzhe Zhu, Yuehan Qin, Li Li, Ziyi Wang, Chaowei Xiao, Yue Zhao

PDF arXiv Code Dataset
NeurIPS 2025 UrbanAI Workshop (NeurIPS 2025 Workshop on Urban AI)

From Image Generation to Infrastructure Design: A Multi-agent Pipeline for Street Design Generation

Chenguang Wang, Xiang Yan, Yilong Dai, Ziyi Wang, Susu Xu

PDF arXiv
arXiv 2025 (arXiv preprint)

Graph Synthetic Out-of-Distribution Exposure with Large Language Models

Haoyan Xu, Zhengtao Yao, Ziyi Wang, Zhan Cheng, Xiyang Hu, Mengyuan Li, Yue Zhao

PDF arXiv
arXiv 2025 (arXiv preprint)

GLIP-OOD: Zero-Shot Graph OOD Detection with Graph Foundation Model

Haoyan Xu, Zhengtao Yao, Xuzhi Zhang, Ziyi Wang, Langzhou He, Yushun Dong, Philip S. Yu, Mengyuan Li, Yue Zhao

PDF arXiv

Background

Research assistant

2026 – presentResearch AssistantTexas A&M University · Dream Lab · Prof. Meng Xia
2025 – 2026Research AssistantUniversity of Southern California · FORTIS Lab · Prof. Yue Zhao
2025 – 2026Research AssistantArizona State University · GLAD Lab · Prof. Xiyang Hu
2025 – 2026Research AssistantUniversity of Florida · Just & Green Transportation Lab · Prof. Xiang Yan
2024 – 2026Research AssistantUniversity of Maryland, College Park · HCIL · Prof. Fumeng Yang & Dr. Zijian Ding

Industry experience

2024AI Product Manager Internbilibili
2023 – 2024HMI UX Design InternBMW Group
2023UX Design InternPublicis Sapient
2023Global Brand Design InternNIO

Service & honors

Program CommitteeAAAI 2026–2027 · AAAI Social Impact Track 2026
Conference ReviewerCHI 2026 Special Recognition · CHI 2027 · ACL 2026 · EMNLP 2026 · NeurIPS 2026 · CSCW 2026 · C&C 2025
Journal ReviewerIJHCI 2026
Student VolunteerDIS 2026
AwardsJenny Preece & Ben Shneiderman Travel Award, University of Maryland, College Park ($1,000)

Teaching

Fall 2025 · Spring 2026Graduate Assistant, INST201 Introduction to Information Science · University of Maryland, College Park

Beyond research

MusicDaft Punk and Kanye West, on repeat.
Cat包子 (Dumpling), a Siamese cat who supervises my writing.
GamesHollow Knight, Celeste and Cuphead.

Ziyi ‘Zoe’ Wang 王子一

News

Publications

Do Vision Language Models Understand Human Engagement in Games?

StreetDesignAI: Broadening Designer Perspectives Through Multi-Persona Evaluation of Cycling Infrastructure

CareerPooler: AI-Powered Metaphorical Pool Simulation Improves Experience and Outcomes in Career Exploration

Persona-aware and Explainable Bikeability Assessment: A Vision-Language Model Approach

Made to Feel: How Designers Bring Emotions into Affective Visualization

Physics-Preserving Latent Compression for Zero-Shot Resolution Transfer in 3D Turbulence

Learning PDE Solvers with Physics and Data: A Unifying View of Physics-Informed Neural Networks and Neural Operators

ResearchCube: Multi-Dimensional Trade-off Exploration for Research Ideation

Sovereign-OS: A Charter-Governed Operating System for Autonomous AI Agents with Verifiable Fiscal Discipline

Multimodal Generative Engine Optimization: Rank Manipulation for Vision-Language Model Rankers

In the Shadow of Judgment: Mapping Out the Landscape of Human–AI Decision-Making Through a Systematic Review

Few-Shot Graph Out-of-Distribution Detection with LLMs

Frontend Diffusion: Empowering Self-Representation of Junior Researchers and Designers Through Multi-agent System

Mitigating Hallucinations in Large Language Models via Causal Reasoning

JailDAM: Jailbreak Detection with Adaptive Memory for Vision-Language Model

From Image Generation to Infrastructure Design: A Multi-agent Pipeline for Street Design Generation

Graph Synthetic Out-of-Distribution Exposure with Large Language Models