Nonverbal cues in text-based computer-mediated communication (CMC), initially introduced to compensate for the lack of social and emotional cues, have evolved beyond their original purpose to express user identity. In particular, embodied identity cues—such as a user’s real face—remain relatively underexplored in text-based CMC despite their potential as richer cues. Recent advances in generative AI have lowered the barrier to AI-mediated self-presentation, yet empirical research is still needed to understand how these cues operate in real interactions and how users experience and accept them. To address this gap, we investigate the social and emotional effects of face-swapped GIFs (FSGIFs) created via generative AI. In a two-phase within-subjects experiment with 32 participants (16 dyads of close acquaintances), we find that FSGIFs significantly enhance relational benefits, including greater co-presence and intimacy compared to generic GIFs. Based on these findings and insights from interviews, we discuss design implications for AI-mediated self-presentation in text-based CMC.

Web agents aim to execute complex online tasks from high-level instructions, yet fully autonomous execution remains challenging in practice. We present an empirical study of user interventions in human–web agent collaboration, moving beyond outcome-based metrics to examine how interventions unfold during execution. We conducted a controlled in-lab study with 30 participants whose interactions reflected early-stage web agent adoption across 12 structured tasks in shopping, travel, and information-seeking domains using live websites. Analyzing interaction logs, user inputs, and screen recordings, we identify diverse behaviors and propose a taxonomy capturing both the reasons for intervention and the forms they take. We distinguish explicit interventions, where users halt or override actions, from implicit interventions, where users guide or prepare the environment without stopping execution. Our findings reveal how task structure and execution breakdowns shape intervention behaviors to provide process-level evidence for designing web agents that better support users as active collaborators.

Abstract Designers often create visualizations to achieve specific high‐level analytical or communication goals. These goals require people to extract complex and interconnected data patterns. Prior perceptual studies of visualization effectiveness have focused on low‐level tasks, such as estimating statistical quantities, and have recently explored high‐level comprehension of visualization. Despite the growing use of Large Language Models (LLMs) as visualization interpreters, how their interpretations relate to human understanding or what reasoning processes underlie their responses remains insufficiently understood. In this work, we explore LLMs' comprehension of visualization, examining the alignment between designers' communicative goals and what their audience sees. We have conducted a qualitative study to investigate the gap between human interpretative strategies and the reasoning pathways of LLMs across three types of visualizations, line graphs, bar graphs, and scatterplots, to identify the high‐level patterns generated by LLMs using three prompt conditions. Our analysis results indicate that LLMs exhibit a consistent interpretative strategy that remains unchanged across prompt constraints. Furthermore, we observe two distinct approaches: humans naturally synthesize data into trend‐centric narratives, whereas LLMs persist with a structural enumeration of comparisons and numerical ranges. Lastly, we see LLMs achieve visualization comprehension through mechanisms distinct from human intuition, pointing to critical challenges and new opportunities for visualization design.

Traffic forecasting represents a crucial problem within intelligent transportation systems.In recent research, Large Language Models (LLMs) have emerged as a promising method, but their intrinsic design, tailored primarily for sequential token processing, introduces notable challenges in effectively capturing spatial dependencies.Specifically, the inherent limitations of LLMs in modeling spatial relationships and their architectural incompatibility with graphstructured spatial data remain largely unaddressed.To overcome these limitations, we introduce ST-LINK, a novel framework that enhances the capability of Large Language Models to capture spatiotemporal dependencies.Its key components are Spatially-Enhanced Attention (SE-Attention) and the Memory Retrieval Feed-Forward Network (MRFFN).SE-Attention extends rotary position embeddings to integrate spatial correlations as direct rotational transformations within the attention mechanism.This approach maximizes spatial learning while preserving the LLM's inherent sequential processing structure.Meanwhile, MRFFN dynamically retrieves and utilizes key historical patterns to capture complex temporal dependencies and improve the stability of long-term forecasting.Comprehensive experiments on benchmark datasets demonstrate that ST-LINK surpasses conventional deep learning and LLM approaches, and effectively captures both regular traffic patterns and abrupt changes.

Directional forecasting in financial markets requires both accuracy and interpretability. Before the advent of deep learning, interpretable approaches based on human-defined patterns were prevalent, but their structural vagueness and scale ambiguity hindered generalization. In contrast, deep learning models can effectively capture complex dynamics, yet often offer limited transparency. To bridge this gap, we propose a two-stage framework that integrates unsupervised pattern extracion with interpretable forecasting. (i) SIMPC segments and clusters multivariate time series, extracting recurrent patterns that are invariant to amplitude scaling and temporal distortion, even under varying window sizes. (ii) JISC-Net is a shapelet-based classifier that uses the initial part of extracted patterns as input and forecasts subsequent partial sequences for short-term directional movement. Experiments on Bitcoin and three S&P 500 equities demonstrate that our method ranks first or second in 11 out of 12 metric--dataset combinations, consistently outperforming baselines. Unlike conventional deep learning models that output buy-or-sell signals without interpretable justification, our approach enables transparent decision-making by revealing the underlying pattern structures that drive predictive outcomes.

Automatically assessing handwritten mathematical solutions is an important problem in educational technology with practical applications, but it remains a significant challenge due to the diverse formats, unstructured layouts, and symbolic complexity of student work.To address this challenge, we introduce VEHME-a Vision-Language Model for Evaluating Handwritten Mathematics Expressions-designed to assess open-form handwritten math responses with high accuracy and interpretable reasoning traces.VEHME integrates a two-phase training pipeline: (i) supervised fine-tuning using structured reasoning data, and (ii) reinforcement learning that aligns model outputs with multi-dimensional grading objectives, including correctness, reasoning depth, and error localization.To enhance spatial understanding, we propose an Expression-Aware Visual Prompting Module, trained on our synthesized multi-line math expressions dataset to robustly guide attention in visually heterogeneous inputs.Evaluated on AIHub and FERMAT datasets, VEHME achieves state-of-the-art performance among open-source models and approaches the accuracy of proprietary systems, demonstrating its potential as a scalable and accessible tool for automated math assessment.Our training and experiment code is publicly available at our GitHub repository.

Comics are an effective method for sequential data-driven storytelling, especially for dynamic graphs-graphs whose vertices and edges change over time. However, manually creating such comics is currently time-consuming, complex, and error-prone. In this paper, we propose DG COMICS, a novel comic authoring tool for dynamic graphs that allows users to semi-automatically build and annotate comics. The tool uses a newly developed hierarchical clustering algorithm to segment consecutive snapshots of dynamic graphs while preserving their chronological order. It also presents rich information on both individuals and communities extracted from dynamic graphs in multiple views, where users can explore dynamic graphs and choose what to tell in comics. For evaluation, we provide an example and report the results of a user study and an expert review.

Accurate traffic forecasting is challenging due to the complex dependency on road networks, various types of roads, and the abrupt speed change due to the events. Recent works mainly focus on dynamic spatial modeling with adaptive graph embedding or graph attention having less consideration for temporal characteristics and in-situ modeling. In this paper, we propose a novel deep learning model named TESTAM, which individually models recurring and non-recurring traffic patterns by a mixture-of-experts model with three experts on temporal modeling, spatio-temporal modeling with static graph, and dynamic spatio-temporal dependency modeling with dynamic graph. By introducing different experts and properly routing them, TESTAM could better model various circumstances, including spatially isolated nodes, highly related nodes, and recurring and non-recurring events. For the proper routing, we reformulate a gating problem into a classification problem with pseudo labels. Experimental results on three public traffic network datasets, METR-LA, PEMS-BAY, and EXPY-TKY, demonstrate that TESTAM achieves a better indication and modeling of recurring and non-recurring traffic.

As online news increasingly include data journalism, there is a corresponding increase in the incorporation of visualization in article thumbnail images. However, little research exists on the design rationale for visualization thumbnails, such as resizing, cropping, simplifying, and embellishing charts that appear within the body of the associated article. Therefore, in this paper we aim to understand these design choices and determine what makes a visualization thumbnail inviting and interpretable. To this end, we first survey visualization thumbnails collected online and discuss visualization thumbnail practices with data journalists and news graphics designers. Based on the survey and discussion results, we then define a design space for visualization thumbnails and conduct a user study with four types of visualization thumbnails derived from the design space. The study results indicate that different chart components play different roles in attracting reader attention and enhancing reader understandability of the visualization thumbnails. We also find various thumbnail design strategies for effectively combining the charts' components, such as a data summary with highlights and data labels, and a visual legend with text labels and Human Recognizable Objects (HROs), into thumbnails. Ultimately, we distill our findings into design implications that allow effective visualization thumbnail designs for data-rich news articles. Our work can thus be seen as a first step toward providing structured guidance on how to design compelling thumbnails for data stories.

With deep learning (DL) outperforming conventional methods for different tasks, much effort has been devoted to utilizing DL in various domains. Researchers and developers in the traffic domain have also designed and improved DL models for forecasting tasks such as estimation of traffic speed and time of arrival. However, there exist many challenges in analyzing DL models due to the black-box property of DL models and complexity of traffic data (i.e., spatio-temporal dependencies). Collaborating with domain experts, we design a visual analytics system, AttnAnalyzer, that enables users to explore how DL models make predictions by allowing effective spatio-temporal dependency analysis. The system incorporates dynamic time warping (DTW) and Granger causality tests for computational spatio-temporal dependency analysis while providing map, table, line chart, and pixel views to assist user to perform dependency and model behavior analysis. For the evaluation, we present three case studies showing how AttnAnalyzer can effectively explore model behaviors and improve model performance in two different road networks. We also provide domain expert feedback.

We present Roslingifier, a data-driven storytelling method for animated scatterplots. Like its namesake, Hans Rosling (1948-2017), a professor of public health and a spellbinding public speaker, Roslingifier turns a sequence of entities changing over time-such as countries and continents with their demographic data-into an engaging narrative elling the story of the data. This data-driven storytelling method with an in-person presenter is a new genre of storytelling technique and has never been studied before. In this article, we aim to define a design space for this new genre-data presentation-and provide a semi-automated authoring tool for helping presenters create quality presentations. From an in-depth analysis of video clips of presentations using interactive visualizations, we derive three specific techniques to achieve this: natural language narratives, visual effects that highlight events, and temporal branching that changes playback time of the animation. Our implementation of the Roslingifier method is capable of identifying and clustering significant movements, automatically generating visual highlighting and a narrative for playback, and enabling the user to customize. From two user studies, we show that Roslingifier allows users to effectively create engaging data stories and the system features help both presenters and viewers find diverse insights.

Music creation is difficult because one must express one's creativity while following strict rules. The advancement of deep learning technologies has diversified the methods to automate complex processes and express creativity in music composition. However, prior research has not paid much attention to exploring the audiences' subjective satisfaction to improve music generation models. In this paper, we evaluate human satisfaction with the state-of-the-art automatic symbolic music generation models using deep learning. In doing so, we define a taxonomy for music generation models and suggest nine subjective evaluation metrics. Through an evaluation study, we obtained more than 700 evaluations from 100 participants, using the suggested metrics. Our evaluation study reveals that the token representation method and models' characteristics affect subjective satisfaction. Through our qualitative analysis, we deepen our understanding of AI-generated music and suggested evaluation metrics. Lastly, we present lessons learned and discuss future research directions of deep learning models for music creation.

Studying history involves many difficult tasks. Examples include searching for proper data in a large event space, understanding stories of historical events by time and space, and finding relationships among events that may not be apparent. Instructors who extensively use well-organized and well-argued materials (e.g., textbooks and online resources) can lead students to a narrow perspective in understanding history and prevent spontaneous investigation of historical events, with the students asking their own questions. In this article, we proposed HisVA, a visual analytics system that allows the efficient exploration of historical events from Wikipedia using three views: event, map, and resource. HisVA provides an effective event exploration space, where users can investigate relationships among historical events by reviewing and linking them in terms of space and time. To evaluate our system, we present two usage scenarios, a user study with a qualitative analysis of user exploration strategies, and in-class deployment results.

To tackle ever-increasing city traffic congestion problems, researchers have proposed deep learning models to aid decision-makers in the traffic control domain. Although the proposed models have been remarkably improved in recent years, there are still questions that need to be answered before deploying models. For example, it is difficult to figure out which models provide state-of-the-art performance, as recently proposed models have often been evaluated with different datasets and experiment environments. It is also difficult to determine which models would work when traffic conditions change abruptly (e.g., rush hour). In this work, we conduct two experiments to answer the two questions. In the first experiment, we conduct an experiment with the state-of-the-art models and the identical public datasets to compare model performance under a consistent experiment environment. We then extract a set of temporal regions in the datasets, whose speeds change abruptly and use these regions to explore model performance with difficult intervals. The experiment results indicate that Graph-WaveNet and GMAN show better performance in general. We also find that prediction models tend to have varying performances with data and intervals, which calls for in-depth analysis of models on difficult intervals for real-world deployment.

As online marketplaces adopt new technologies to encourage consumers’ purchases (e.g., one-click purchases), the number of consumers who impulsively buy products also increases. Although some interventions have been introduced for consumers’ self-controlled purchases, there have been few studies that evaluate the effectiveness of the techniques in the real environment. In this paper, we conducted an online survey with 118 consumers in their 20s to investigate their impulse buying behaviors and self-control strategies. Based on the survey results and literature surveys, we developed interventions that can assist consumers in controlling their online purchase habits, including Reflection, Distraction, Desire Reduction, and Salient Cost. For evaluation, we enrolled 107 participants in a user study on a real-world e-commerce site. The results indicate that all interventions were effective in reducing impulse buying urges, with variations in user experiences. Our findings and design implications are discussed.

Users may face challenges while designing graphical user interfaces, due to a lack of relevant experience and guidance. This paper aims to investigate the issues users face during the design process, and how to resolve them. To this end, we conducted semi-structured interviews, based on which we built a GUI prototyping assistance tool called GUIComp. This tool can be connected to GUI design software as an extension, and it provides real-time, multi-faceted feedback on a user's current design. Additionally, we conducted two user studies, in which we asked participants to create mobile GUIs with or without GUIComp, and requested online workers to assess the created GUIs. The experimental results show that GUIComp facilitated iterative designs and the participants with GUIComp had better a user experience and produced more acceptable designs than those who did not use it.

Predicting road traffic speed is a challenging task due to different types of roads, abrupt speed change and spatial dependencies between roads; it requires the modeling of dynamically changing spatial dependencies among roads and temporal patterns over long input sequences. This paper proposes a novel spatio-temporal graph attention (ST-GRAT) that effectively captures the spatio-temporal dynamics in road networks. The novel aspects of our approach mainly include spatial attention, temporal attention, and spatial sentinel vectors. The spatial attention takes the graph structure information (e.g., distance between roads) and dynamically adjusts spatial correlation based on road states. The temporal attention is responsible for capturing traffic speed changes, and the sentinel vectors allow the model to retrieve new features from spatially correlated nodes or preserve existing features. The experimental results show that ST-GRAT outperforms existing models, especially in difficult conditions where traffic speeds rapidly change (e.g., rush hours). We additionally provide a qualitative study to analyze when and where ST-GRAT tended to make accurate predictions during rush-hour times.

We present an interactive visual analytics system that enables traffic congestion exploration, surveillance, and forecasting based on vehicle detector data. Through domain expert collaboration, we have extracted task requirements, incorporated the Long Short-Term Memory (LSTM) model for congestion forecasting, and designed a weighting method for detecting the causes of congestion and congestion propagation directions. Our visual analytics system is designed to enable users to explore congestion causes, directions, and severity. Congestion conditions of a city are visualized using a Volume-Speed Rivers (VSRivers) visualization that simultaneously presents traffic volumes and speeds. To evaluate our system, we report performance comparison results, wherein our model is more accurate than other forecasting algorithms. We demonstrate the usefulness of our system in the traffic management and congestion broadcasting domains through three case studies and domain expert feedback.

In spite of the evolution of Augmented Reality~(AR) technology, it is not wide spread in everyday life. There may be many reasons, but one of the reasons is that it has been developed for very specific users, such as researchers and professionals. To overcome this problem, Grubert et al. proposed the pervasive AR. It is not limited to a specific situation, but is usable in various instances and providing continuous and flexible AR experience. The AR browser is the example of utilizing pervasive AR. The AR browser understands the context of the user and provides corresponding information. However, if the corresponding information to the context of user cannot reach the user in time due to massive data transmission, unexpected network congestion, poor service quality or signal strength, it cannot be guaranteed to be continuous. This leads to a degradation of the user experience, and it cannot support pervasive AR. This paper presents the Progressive Augmented Reality, the way which quickly send incomplete, yet informative, response about the user's current context rather than wait for sending complete information to the user. The concept of Progressive Augmented Reality comes from Progressive Data Science. Our system is aware of network quality by collecting various network health parameters. According to the network status quality, it divides the chunk information to the optimal number and transmits the one that have the highest priority among the divided information. By conducting the above process iteratively, all the divided information is updated. Our client-side system utilizes Android ARCore and has been tested on Google Pixel 2XL.

Addiction is a brain disease manifested by the loss of control over drugs or behaviors, despite negative consequences. Although addiction research has been conducted for decades in psychiatry and neuroscience, a comprehensive understanding of the mechanisms underlying addiction has not yet been achieved. Recent studies in neuroscience [1] have sought to bring light upon this issue by measuring exploration/exploitation decisions in sequential choice tasks, requiring balancing the need to exploit known options and to explore new ones. These studies show a relationship between addiction and exploration/exploitation decisions. For example, people addicted to substances (e.g. alcohol or methamphetamine) or behaviors (e.g. gambling) have tendencies to explore less, which implies they have difficulties 'seeing the big picture".

Abstract Market participants and businesses have made tremendous efforts to make the best decisions in a timely manner under varying economic and business circumstances. As such, decision‐making processes based on Financial data have been a popular topic in industries. However, analyzing Financial data is a non‐trivial task due to large volume, diversity and complexity, and this has led to rapid research and development of visualizations and visual analytics systems for Financial data exploration. Often, the development of such systems requires researchers to collaborate with Financial domain experts to better extract requirements and challenges in their tasks. Work to systematically study and gather the task requirements and to acquire an overview of existing visualizations and visual analytics systems that have been applied in Financial domains with respect to real‐world data sets has not been completed. To this end, we perform a comprehensive survey of visualizations and visual analytics. In this work, we categorize Financial systems in terms of data sources, applied automated techniques, visualization techniques, interaction, and evaluation methods. For the categorization and characterization, we utilize existing taxonomies of visualization and interaction. In addition, we present task requirements extracted from interviews with domain experts in order to help researchers design better systems with detailed goals.

Abstract Real‐time microblogs can be utilized to provide situational awareness during emergency and disaster events. However, the utilization of these datasets requires the decision makers to perform their exploration and analysis across a range of data scales from local to global, while maintaining a cohesive thematic context of the transition between the different granularity levels. The exploration of different information dimensions at the varied data and human scales remains to be a non‐trivial task. To this end, we present a visual analytics situational awareness environment that supports the real‐time exploration of microblog data across multiple scales of analysis. We classify microblogs based on a fine‐grained, crisis‐related categorization approach, and visualize the spatiotemporal evolution of multiple categories by coupling a spatial lens with a glyph‐based visual design. We propose a transparency‐based spatial context preserving technique that maintains a smooth transition between different spatial scales. To evaluate our system, we conduct user studies and provide domain expert feedback.

We present VASA, a visual analytics platform consisting of a desktop application, a component model, and a suite of distributed simulation components for modeling the impact of societal threats such as weather, food contamination, and traffic on critical infrastructure such as supply chains, road networks, and power grids. Each component encapsulates a high-fidelity simulation model that together form an asynchronous simulation pipeline: a system of systems of individual simulations with a common data and parameter exchange format. At the heart of VASA is the Workbench, a visual analytics application providing three distinct features: (1) low-fidelity approximations of the distributed simulation components using local simulation proxies to enable analysts to interactively configure a simulation run; (2) computational steering mechanisms to manage the execution of individual simulation components; and (3) spatiotemporal and interactive methods to explore the combined results of a simulation run. We showcase the utility of the platform using examples involving supply chains during a hurricane as well as food contamination in a fast food restaurant chain.

The concept of preconditioning data (utilizing a power transformation as an initial step) for analysis and visualization is well established within the statistical community and is employed as part of statistical modeling and analysis. Such transformations condition the data to various inherent assumptions of statistical inference procedures, as well as making the data more symmetric and easier to visualize and interpret. In this paper, we explore the use of the Box-Cox family of power transformations to semiautomatically adjust visual parameters. We focus on time-series scaling, axis transformations, and color binning for choropleth maps. We illustrate the usage of this transformation through various examples, and discuss the value and some issues in semiautomatically using these transformations for more effective data visualization.

Abstract Competitive intelligence is a systematic approach for gathering, analyzing, and managing information to make informed business decisions. Many companies use competitive intelligence to identify risks and opportunities within markets. Point of sale data that retailers share with vendors is of critical importance in developing competitive intelligence. However, existing tools do not easily enable the analysis of such large and complex data. therefore, new approaches are needed in order to facilitate better analysis and decision making. In this paper, we present MarketAnalyzer, an interactive visual analytics system designed to allow vendors to increase their competitive intelligence. MarketAnalyzer utilizes pixel‐based matrices to present sale data, trends, and market share growths of products of the entire market within a single display. These matrices are augmented by advanced underlying analytical methods to enable the quick evaluation of growth and risk within market sectors. Furthermore, our system enables the aggregation of point of sale data in geographical views that provide analysts with the ability to explore the impact of regional demographics and trends. Additionally, overview and detailed information is provided through a series of coordinated multiple views. In order to demonstrate the effectiveness of our system, we provide two use‐case scenarios as well as feedback from market analysts.