Publications

PacificVis · Extended Reality

LandSAR: Visceralizing Landslide Data for Enhanced Situational Awareness in Immersive Analytics

Wong Kam-Kwai, Yi-Lin Ye, Wai Tong, Haobo Li, Kentaro Takahira, Aastha Bhatta, Sunil Poudyal, Charles Wang Wai Ng, Huamin Qu, and Leni Yang

Lets planners rehearse landslide scenarios on a physical terrain model before disasters strike.How it works

Pairs real-time what-if landslide simulation with a tangible terrain interface, turning hazard assessment into prospective spatial reasoning for mitigation and situational awareness.

UIST · Extended Reality

RegulAR: Graph-Grounded Error Recognition and Assistance for Procedural Tasks in Augmented Reality

Yi-Lin Ye, Jindu Wang, Hiu Tung Wong, Shuchang Xu, Huamin Qu, and Wong Kam-Kwai (corresponding author)

Helps people understand what went wrong, why it matters, and how to recover during procedural tasks.How it works

Grounds a multimodal large language model in a hierarchical task-dependency graph so an AR assistant can track procedural progress, classify deviations, estimate downstream impact, and provide situated recovery guidance.

To appear

SUI · Extended Reality

Surrounded by Friends: Design and Evaluation of Immersive Layouts of Egocentric Network for Visual Analytics

Kentaro Takahira, Takanori Fujiwara, Wong Kam-Kwai, Kento Shigyo, Leni Yang, Hiroaki Natsukawa, Yalong Yang, and Huamin Qu

Shows which immersive network layouts best support connection-strength and topology analysis.How it works

Compares four immersive egocentric-network layouts across connection-strength and topology tasks, showing that Cube favors ego-alter strength judgments while Spherical improves topology understanding and reduces occlusion.

FCS · Extended Reality

ARtiVision: Bridging Expert Knowledge and Visitor Experience through Gaze-Guided Artifact Interpretation in AR

Wei Zhang, Xing Liu, Biying XU, Xinzhuo DENG, WONG Kam-Kwai, Wenjie NING, and Wei CHEN

Provides personalized, in-the-moment guidance to museum visitors for deeper understanding.How it works

A gaze-guided AR interpretation framework that encodes curatorial knowledge in a hierarchical appreciation model and reveals layered, preference-aware explanations based on what visitors attend to and how deeply they want to engage.

TVCG · Business Intelligence

Prismatic: Interactive Multi-View Cluster Analysis of Concept Stocks

Wong Kam-Kwai, Yan Luo, Xuanwu Yue, Wei Chen, and Huamin Qu

Helps investors see why stocks belong together instead of trusting correlations alone.How it works

Couples data-driven co-movement with explicit business relations so analysts can steer concept-stock generation, exploration, and validation across time horizons.

ACM MM · Digital Humanities

CultiVerse: Towards Cross-Cultural Understanding for Paintings with Large Language Model

Wei Zhang, Wong Kam-Kwai, Biying Xu, Yiwen Ren, Yuhuai Li, Yingchaojie Feng, Minfeng Zhu, and Wei Chen

Makes culturally specific symbolism in paintings easier for outsiders to understand.How it works

Builds cross-cultural intelligibility through a mixed-initiative workflow and cultural-norm dataset that align symbols across cultures, surface context across visual/symbol/custom levels, and validate acceptable interpretations.

ACM CHI · Extended Reality

TangibleNet: Synchronous Network Data Storytelling through Tangible Interactions in Augmented Reality

Kentaro Takahira, Wong Kam-Kwai, Leni Yang, Xian Xu, Takanori Fujiwara, and Huamin Qu

Gives live presenters a more natural way to manipulate network stories in space.How it works

Maps tangible AR actions to node-link operations so presenters can steer synchronous network storytelling in real time without breaking presentation flow.

TVCG · LLM

Save It for the “Hot” Day: An LLM-Empowered Visual Analytics System for Heat Risk Management

Haobo Li, Wong Kam-Kwai, Yan Luo, Juntong Chen, Chengzhong Liu, Yaxuan Zhang, Alexis Kai Hon Lau, Huamin Qu, and Dongyu Liu

Helps heat-risk planning account for social context, not just weather forecasts.How it works

Fuses numerical heat models with LLM-structured news evidence, creating a prospective decision space where experts can retrieve, align, and act on contextual risk signals for mitigation.

ACM CHI · Extended Reality

Data Bias Recognition in Museum Settings: Framework Development and Contributing Factors

Stella Quinto Lima, Gabriela Buraglia, Wong Kam-Kwai, and Jessica Roberts

Shows how data exhibits can teach visitors to question bias instead of passively trust them.How it works

Identifies familiarity as a key enabler of bias recognition and formalizes a three-layer framework for designing more critically intelligible AR data exhibitions.

ACM CHI · Extended Reality

Ego vs. Exo and Active vs. Passive: Investigating the Individual and Combined Effects of Viewpoint and Navigation on Spatial Immersion and Understanding in Immersive Storytelling

Tao Lu, Qian Zhu, Tiffany Ma, Wong Kam-Kwai, Anlan Xie, Alex Endert, and Yalong Yang

Shows when immersive stories should give audiences control or guide them.How it works

Separates the effects of viewpoint and navigation in immersive storytelling, showing how egocentric+active boosts agency while exocentric+passive preserves focus on story content.

TVCG · Extended Reality

Exploring Spatial Hybrid User Interface for Visual Sensemaking

Wai Tong, Haobo Li, Meng Xia, Wong Kam-Kwai, Ting-Chuen Pong, Huamin Qu, and Yalong Yang

Shows that combining desktop precision with VR space can improve analytic work without costly switching.How it works

Distributes visual sensemaking across PC and VR, letting users combine desktop precision with spatial overview rather than forcing one device to do both.

C&G · Information Visualization

CausalPrism: A Visual Analytics Approach for Subgroup-based Causal Heterogeneity Exploration

Xingyu Liu, Jiehui Zhou, Xumeng Wang, Wong Kam-Kwai, Wei Zhang, Xingyu Liu, Juntian Zhang, Minfeng Zhu, and Wei Chen

Helps researchers see who benefits from a treatment and how reliable that pattern is.How it works

Combines Pareto-optimized, rule-explained subgroup discovery with interactive ranking and uncertainty views, turning subgroup HTE exploration into a steerable trade-off space.

VI · Business Intelligence

FundSelector: A visual analysis system for mutual fund selection

Fan Yan, Yong Wang, Xuanwu Yue, Wong Kam-Kwai, Ketian Mao, Rong Zhang, Huamin Qu, Haiyang Zhu, Minfeng Zhu, and Wei Chen

Helps investors choose funds that match their priorities rather than generic rankings.How it works

Quantifies investor preferences and exposes factor-level reasons and temporal trade-offs behind each ranking, making mutual fund selection steerable rather than one-size-fits-all.

PacificVis · Digital Humanities

ScrollTimes: Tracing the Provenance of Paintings as a Window into History

Wei Zhang, Wong Kam-Kwai, Yitian Chen, Ailing Jia, Luwei Wang, Jian-Wei Zhang, Lechao Cheng, Huamin Qu, and Wei Chen

Turns a painting's scattered traces into a readable life history.How it works

Structures provenance as a three-level biography workflow linking extracted visual evidence, validated historical context, and temporal transmission into customizable handscroll histories.

JCST · Digital Humanities

Computational Approaches for Traditional Chinese Painting: From the “Six Principles of Painting” Perspective

Wei Zhang, Jianwei Zhang, Wong Kam-Kwai, Yifang Wang, Yingchaojie Feng, Luwei Wang, and Wei Chen

Shows where computing has and has not yet engaged the core ideas of Chinese painting.How it works

Re-reads 92 papers through the Six Principles and application stages, turning computational TCP into a prospecting map of underexplored artistic dimensions.

Info Vis · LLM

Towards an Understanding and Explanation for Mixed-Initiative Artificial Scientific Text Detection

Luoxuan Weng, Shi Liu, Hang Zhu, Jiashun Sun, Wong Kam-Kwai, Dongming Han, Minfeng Zhu, and Wei Chen

Makes AI-text detection more trustworthy by showing what cues drive the judgment.How it works

Combines human priors, model evidence, and contribution-based explanations in a mixed-initiative workflow that makes scientific-text detection more intelligible and trustworthy.

TVCG · Business Intelligence

FMLens: Towards Better Scaffolding the Process of Fund Manager Selection in Fund Investments

Longfei Chen, Chen Cheng, He Wang, Xiyuan Wang, Yun Tian, Xuanwu Yue, Wong Kam-Kwai, Haipeng Zhang, Suting Hong, and Quan Li

Helps investors judge fund managers on style and skill instead of headline returns alone.How it works

Decomposes fund managers' styles, abilities, and performance persistence into linked views that make selection more intelligible than comparing headline returns alone.

TVCG · Business Intelligence

Anchorage: Visual Analysis of Satisfaction in Customer Service Videos via Anchor Events

Wong Kam-Kwai, Xingbo Wang, Yong Wang, Jianben He, Rong Zhang, and Huamin Qu

Turns long service videos into readable evidence for where satisfaction breaks down.How it works

Anchors multimodal behavior to semantically meaningful service events, letting analysts navigate long videos directly through operational causes of satisfaction change.

IEEE VIS · LLM

PromptMagician: Interactive Prompt Engineering for Text-to-Image Creation

Yingchaojie Feng, Xingbo Wang, Wong Kam-Kwai, Sijia Wang, Yuhong Lu, Minfeng Zhu, Baicheng Wang, and Wei Chen

Helps novices turn vague image ideas into prompts they can deliberately refine.How it works

Retrieves similar prompt-image examples, surfaces influential keywords, and visualizes cross-modal neighborhoods so users can steer prompt refinement for text-to-image creation.

IEEE VR · Extended Reality

Towards an Understanding of Distributed Asymmetric Collaborative Visualization on Problem-solving

Wai Tong, Meng Xia, Wong Kam-Kwai, Doug A Bowman, Ting-Chuen Pong, Huamin Qu, and Yalong Yang

Shows mixed PC-VR teams can work effectively without everyone using the same device.How it works

Compares asymmetric PC-VR with symmetric setups and iterates the interface to show how distributed teams can steer role split, effort, and coordination across devices.

IEEE VIS · Information Visualization

VideoPro: A Visual Analytics Approach for Interactive Video Programming

Jianben He, Xingbo Wang, Wong Kam-Kwai, Xijie Huang, Changjian Chen, Zixin Chen, Fengjie Wang, Min Zhu, and Huamin Qu

Lets experts build video labels faster while keeping control over what the model learns.How it works

Turns detected video events into reusable labeling-function templates and exposes their downstream effect on model performance, enabling steerable video data programming.

VI · Business Intelligence

Tax-Scheduler: An Interactive Visualization System for Staff Shifting and Scheduling at Tax Authorities

Linping Yuan, Boyu Li, Siqi Li, Wong Kam-Kwai, Rong Zhang, and Huamin Qu

Helps tax offices plan staffing ahead of demand instead of reacting after queues form.How it works

Couples client-inflow prediction with interactive staff-scheduling optimization, turning historical demand into prospective staffing plans for tax service operations.

TVCG · LLM

XNLI: Explaining and Diagnosing NLI-based Visual Data Analysis

Yingchaojie Feng, Xingbo Wang, Bo Pan, Wong Kam-Kwai, Yi Ren, Shi Liu, Zihan Yan, Yuxin Ma, Huamin Qu, and Wei Chen

Shows users why a natural-language chart request went wrong and how to fix it.How it works

Exposes provenance across attributes, tasks, and encodings, then couples those explanations with interactive fixes and query-revision hints for intelligible NLI analysis.

ChinaVis · Business Intelligence

Visual Analytics for Security Threats Detection in Ethereum Consensus Layer

Xuan Chen, Xincan Zhang, Zhaohan Wang, Kerun Yu, Wong Kam-Kwai, Haoyun Guo, and Siming Chen

Makes hidden beacon-chain attacks easier for developers to notice and investigate.How it works

Links system-level security state to slot-level consensus outcomes and attack-specific anomalies, making Ethereum consensus threats intelligible to developers.

IEEE VIS · Digital Humanities

CohortVA: A Visual Analytic System for Interactive Exploration of Cohorts based on Historical Data

Wei Zhang, Wong Kam-Kwai, Xumeng Wang, Youcheng Gong, Rongchen Zhu, Kai Liu, Zihan Yan, Siwei Tan, Huamin Qu, Siming Chen, and Wei Chen

Helps historians test and refine groupings of historical figures instead of accepting opaque mining results.How it works

Combines knowledge-graph candidates with interactive views that let historians authenticate figures, compare cohort features, and iterate hypotheses across space and time.

IEEE VIS · Information Visualization

DPVisCreator: Incorporating Pattern Constraints to Privacy-preserving Visualizations via Differential Privacy

Jiehui Zhou, Xumeng Wang, Wong Kam-Kwai, Huanliang Wang, Zhongwei Wang, Xiaoyu Yang, Xiaoran Yan, Haozhe Feng, Huamin Qu, Haochao Ying, and Wei Chen

Creates privacy-preserving charts without blindly destroying the patterns they care about.How it works

Injects user-specified pattern constraints into a differential-privacy model so privacy-preserving charts remain steerable around the visual patterns users care about.

TVCG · Information Visualization

Explaining with Examples Lessons Learned from Crowdsourced Introductory Description of Information Visualizations

Leni Yang, Cindy Xiong, Wong Kam-Kwai, Aoyu Wu, and Huamin Qu

Shows that concrete examples help people learn to read charts.How it works

Validates example-based introduction strategies as an effective way to teach visual encodings, making chart explanations more intelligible than abstract descriptions alone.

ChinaVis · Digital Humanities

iPoet: Interactive Painting Poetry Creation with Visual Multimodal Analysis

Yingchaojie Feng, Jiazhou Chen, Keyu Huang, Wong Kam-Kwai, Hui Ye, Wei Zhang, Rongchen Zhu, Xiaonan Luo, and Wei Chen

Lets non-experts co-create poems that respond to what a painting depicts and expresses.How it works

Grounds poetry generation in extracted visual content and inferred emotion, then lets users guide the resulting poetic interpretation of ancient paintings.

IEEE VIS · Business Intelligence

TaxThemis: Interactive Mining and Exploration of Suspicious Tax Evasion Groups

Yating Lin (equal contribution), Wong Kam-Kwai (equal contribution), Yong Wang, Rong Zhang, Bo Dong, Huamin Qu, and Qinghua Zheng

Makes group tax-evasion schemes visible as connected patterns rather than isolated suspicious transactions.How it works

Fuses ownership and trade networks with evidence views that expose suspicious group structure, profit flow, and transfer patterns in related-party tax evasion.