Fine-grained Text-Visualization Integration:Enhancing Data Communication

The Hong Kong University of Science and Technology (Guangzhou)

Data Science and Analytics Thrust

PhD Qualifying Examination

By Mr. WANG Liangwei

Abstract

Effective data communication hinges on the seamless integration of textual descriptions and visual representations. While numerous methods exist, a systematic understanding of fine-grained, element level coordination remains underdeveloped. This survey proposes a structured framework to analyze Visualization-Text Interplay (VTI) across three key paradigms: semantic linking, visualization-grounded text generation, and language-driven visualization synthesis. Through a 4W1H analytical lens, we review 33 representative systems from top-tier venues, examining their user motivations, scenarios, construction workflows, interaction mechanisms, and implementation strategies. Our findings reveal recurring design patterns and technical gaps that hinder broader adoption. We conclude by outlining directions for future research, calling for context-aware frameworks and multimodal models that can reason about semantic correspondences across modalities with greater flexibility and fidelity

PQE Committee

Chair of Committee: Prof. TANG Nan

Prime Supervisor: Prof. TSUNG Fugee

Co-Supervisor: Prof. LUO Yuyu

Examiner: Prof. YANG Weikai