GAMES Webinar 2025 – 366期(文本深度理解中的可视设计与交互) | 邱瑞(俄亥俄州立大学)，殷佳宁(浙江大学)
【GAMES Webinar 2025-366期】(可视化专题-文本深度理解中的可视设计与交互)
报告嘉宾：邱瑞（俄亥俄州立大学）
报告时间：2025年5月15号星期四晚上8:00-8:30（北京时间）
报告题目：VADIS: A Visual Analytics Pipeline for Dynamic Document Representation and Information-Seeking
报告摘要：
In the biomedical domain, visualizing the document embeddings of an extensive corpus has been widely used in information-seeking tasks. However, three key challenges with existing visualizations make it difficult for clinicians to find information efficiently. First, the document embeddings used in these visualizations are generated statically by pretrained language models, which cannot adapt to the user’s evolving interest. Second, existing document visualization techniques cannot effectively display how the documents are relevant to users’ interest, making it difficult for users to identify the most pertinent information. Third, existing embedding generation and visualization processes suffer from a lack of interpretability, making it difficult to understand, trust and use the result for decision-making. In this paper, we present a novel visual analytics pipeline for user driven document representation and iterative information seeking (VADIS). VADIS introduces a prompt-based attention model (PAM) that generates dynamic document embedding and document relevance adjusted to the user’s query. To effectively visualize these two pieces of information, we design a new document map that leverages a circular grid layout to display documents based on both their relevance to the query and the semantic similarity. Additionally, to improve the interpretability, we introduce a corpus-level attention visualization method to improve the user’s understanding of the model focus and to enable the users to identify potential oversight. This visualization, in turn, empowers users to refine, update and introduce new queries, thereby facilitating a dynamic and iterative information-seeking experience.
讲者简介：
Rui Qiu (邱瑞) is a fourth-year Ph.D. student in the Gravity Lab at The Ohio State University, advised by Prof. Han-Wei Shen. His research explores text visualization and human-centered design, with a current emphasis on augmenting agentic workflows—investigating how nuanced human interaction and oversight can enhance both system performance and interpretability.
讲者主页：https://ryanq96.github.io/
报告嘉宾：殷佳宁(浙江大学)
报告时间：2025年5月15号星期四晚上8:30-8:50（北京时间）
报告题目：Blowing Seeds across Gardens: Visualizing Implicit Propagation of Cross-Platform Social Media Posts
报告摘要：
Propagation analysis refers to studying how information spreads on social media, a pivotal endeavor for understanding social sentiment and public opinions. Numerous studies contribute to visualizing information spread, but few have considered the implicit and complex diffusion patterns among multiple platforms. To bridge the gap, we summarize cross-platform diffusion patterns with experts and identify significant factors that dissect the mechanisms of cross-platform information spread. Based on that, we propose an information diffusion model that estimates the likelihood of a topic/post spreading among different social media platforms. Moreover, we propose a novel visual metaphor that encapsulates cross-platform propagation in a manner analogous to the spread of seeds across gardens. Specifically, we visualize platforms, posts, implicit cross-platform routes, and salient instances as elements of a virtual ecosystem — gardens, flowers, winds, and seeds, respectively. We further develop a visual analytic system, namely BloomWind, that enables users to quickly identify the cross-platform diffusion patterns and investigate the relevant social media posts. Ultimately, we demonstrate the usage of BloomWind through two case studies and validate its effectiveness using expert interviews.
讲者简介：
Jianing Yin (殷佳宁) is a second-year Ph.D. candidate at the State Key Lab of CAD&CG, Zhejiang University, supervised by Prof. Yingcai Wu (巫英才) and Dr. Tan Tang (唐谈). Her research lies at the intersection of data visualization and human-computer interaction, with a particular focus on analyzing and understanding social media ecosystems. She explores how visual metaphors and interactive techniques can be used to reveal diverse phenomena and patterns across and within social media platforms.
主持人简介：
Chengtao Ji received his M.S. and Doctoral degree from Sichuan University and Groningen University in 2014 and 2018, respectively. He is currently an assistant professor in Xi’an Jiaotong-Liverpool University, China. His research interests include data mining, complex network analysis, and visual analytics.
Yunzhe Wang received her M.S. and PhD in Computer Science from The University of Hong Kong and The Hong Kong Polytechnic University in 2014 and 2020, respectively. She is currently a lecturer in Suzhou University of Science and Technology, China. Her research interests include data science, social networks, and visual analysis. She has published paper on top conferences and journals including Computer Graphics Forum and ACM Transactions on Knowledge Discovery from Data and so on. She received ICCI*CC Best Paper Award in 2015, SIGGRAPH Asia Sym. Vis. Best Paper Award in 2017 and Jiangsu Innovation and Entrepreneurship Ph.D. from World Prestigious Universities Award in 2021. She now hosts a project supported by the Young Scientists Fund of the National Natural Science Foundation of China.
GAMES主页的“使用教程”中有 “如何观看GAMES Webinar直播？”及“如何加入GAMES微信群？”的信息；
GAMES主页的“资源分享”有往届的直播讲座的视频及PPT等。
观看直播的链接：https://live.bilibili.com/h5/24617282
