Large language models (LLMs) are increasingly applied in specialized domains such as finance and healthcare, where they introduce unique safety risks.Domain-specific datasets of harmful prompts remain scarce and still largely rely on manual construction; public datasets mainly focus on explicit harmful prompts, which modern LLM defenses can often detect and refuse.In contrast, implicit harmful prompts-expressed through indirect domain knowledge-are harder to detect and better reflect real-world threats.We identify two challenges: transforming domain knowledge into actionable constraints and increasing the implicitness of generated harmful prompts.To address them, we propose an end-to-end framework that first performs knowledge-graphguided harmful prompt generation to systematically produce domain-relevant prompts, and then applies two-strategy obfuscation rewriting to convert explicit harmful prompts into implicit variants via direct and context-enhanced rewriting.This framework yields high-quality datasets combining strong domain relevance with implicitness, enabling more realistic redteaming and advancing LLM safety research.We release our code and datasets on GitHub.

Chart data extraction, which reverse-engineers data tables from chart images, is essential for reproducibility, analysis, retrieval, and redesign. Existing interactive tools are reliable but tedious, and mixed-initiative systems, while more efficient, lack generalizability. Recent multimodal large language models (MLLMs) offer a unified interface for chart interpretation, yet their ability to extract accurate data tables, especially without visible labels, remains unclear. We build a benchmark featuring diverse real-world charts without data labels to evaluate this capability. Results show that, while current MLLMs reliably reconstruct table structures, they struggle with precise value recovery. To address this, we revisit chart data extraction from a human-centered perspective and argue that extraction should follow a progressive learning process similar to how people read charts. Our training framework substantially improves numerical accuracy, achieving state-of-the-art performance with a 7B-parameter model. A user study further shows that our model effectively supports mixed-initiative workflows for reliable chart data extraction.

Computational notebooks offer a flexible environment for exploratory data analysis (EDA), but this flexibility often leads to disorganized and iterative execution of notebook cells, making it difficult to track how data states evolve. Consequently, data scientists must devote extra mental effort to staying aware of data states, which is both tedious and prone to overlooking anomalies. To address this challenge, we developed NoteFlow, a notebook extension that leverages charts as “sight glasses” to provide a consistent and continuous tracing of data flow. NoteFlow allows users to (1) validate various facets of the current data state using recommended charts provided immediately after each cell execution, and (2) trace the global evolution of selected charts to continuously observe how particular data attributes evolve throughout the EDA process. We evaluated NoteFlow’s effectiveness through a controlled study with 12 participants and a one-month field study with 2 data scientists on real-world workflows.

Large Language Models (LLMs) demonstrate exceptional capabilities in factual question answering, yet they sometimes provide incorrect responses. To address this issue, knowledge editing techniques have emerged as effective methods for correcting factual information in LLMs. However, typical knowledge editing workflows struggle with identifying the optimal set of model layers for editing and rely on summary indicators that provide insufficient guidance. This lack of transparency hinders effective comparison and identification of optimal editing strategies. In this paper, we present KEditVis, a novel visual analytics system designed to assist users in gaining a deeper understanding of knowledge editing through interactive visualizations, improving editing outcomes, and discovering valuable insights for the future development of knowledge editing algorithms. With KEditVis, users can select appropriate layers as the editing target, explore the reasons behind ineffective edits, and perform more targeted and effective edits. Our evaluation, including usage scenarios, expert interviews, and a user study, validates the effectiveness and usability of the system.

In basketball, decision-making is one of the core skills for players. For example, when a player is holding the ball, the success of the team’s offense is primarily determined by her/his decisions (i.e., pass, shoot, or dribble) in response to the dynamics of the game. Understanding players’ decision-making processes in changing game situations can help coaches develop effective strategies, which is critical for the success of a team. However, the decision-making process is influenced by various factors (e.g., player’s playing style, opponents’ defense, and time remaining), making understanding a challenging problem. In this study, we propose HoopScouter, a visual analytics system to help understand ball handlers’ decisions in basketball games. Based on a careful investigation of the analysis requirements, we first introduce a representation learning method that characterizes ball handlers’ decision-making styles. We then design a sketch panel with integrated time information to support exploration of player decisions under similar game scenarios. Facet views and coordinated interactions are also provided to identify the strengths and weaknesses of the ball handler’s decision-making, and to understand when and why ball handlers would make certain decisions. To validate the effectiveness of HoopScouter, we conduct two case studies on real-world basketball games and receive positive feedback from domain experts.

Figure 1: ReSpark Overview.ReSpark helps users generate new data reports by reusing existing ones.It begins by ranking reports based on their alignment with the target dataset to assist in selecting a reference report (a).The selected report is then segmented into a sequence of analysis segments (b).For each segment, ReSpark extracts the analysis objective (c) and adapts it to the target dataset (d1).It then generates the corresponding code and charts (d2), followed by textual insights (d3).These elements are finally composed into a new report.

To enhance the appeal and informativeness of data news, there is an increasing reliance on data analysis techniques and visualizations, which poses a high demand for journalists' abilities. While numerous visual analytics systems have been developed for deriving insights, few tools specifically support and disseminate viewpoints for journalism. Thus, this work aims to facilitate the automatic creation of sports news from natural language insights. To achieve this, we conducted an extensive preliminary study on the published sports articles. Based on our findings, we propose a workflow - 1) exploring the data space behind insights, 2) generating narrative structures, 3) progressively generating each episode, and 4) mapping data spaces into communicative visualizations. We have implemented a human-AI interaction system called SNIL, which incorporates user input in conjunction with large language models (LLMs). It supports the modification of textual and graphical content within the episode-based structure by adjusting the description. We conduct user studies to demonstrate the usability of SNIL and the benefit of bridging the gap between analysis tasks and communicative tasks through expert and fan feedback.

Large Language Models (LLMs) are powerful but also raise significant security concerns, particularly regarding the harm they can cause, such as generating fake news that manipulates public opinion on social media and providing responses to unethical activities. Traditional red teaming approaches for identifying AI vulnerabilities rely on manual prompt construction and expertise. This paper introduces AdversaFlow, a novel visual analytics system designed to enhance LLM security against adversarial attacks through human-AI collaboration. AdversaFlow involves adversarial training between a target model and a red model, featuring unique multi-level adversarial flow and fluctuation path visualizations. These features provide insights into adversarial dynamics and LLM robustness, enabling experts to identify and mitigate vulnerabilities effectively. We present quantitative evaluations and case studies validating our system's utility and offering insights for future AI security solutions. Our method can enhance LLM security, supporting downstream scenarios like social media regulation by enabling more effective detection, monitoring, and mitigation of harmful content and behaviors.

The use of natural language interfaces (NLIs) to create charts is becoming increasingly popular due to the intuitiveness of natural language interactions. One key challenge in this approach is to accurately capture user intents and transform them to proper chart specifications. This obstructs the wide use of NLI in chart generation, as users' natural language inputs are generally abstract (i.e., ambiguous or under-specified), without a clear specification of visual encodings. Recently, pre-trained large language models (LLMs) have exhibited superior performance in understanding and generating natural language, demonstrating great potential for downstream tasks. Inspired by this major trend, we propose ChartGPT, generating charts from abstract natural language inputs. However, LLMs are struggling to address complex logic problems. To enable the model to accurately specify the complex parameters and perform operations in chart generation, we decompose the generation process into a step-by-step reasoning pipeline, so that the model only needs to reason a single and specific sub-task during each run. Moreover, LLMs are pre-trained on general datasets, which might be biased for the task of chart generation. To provide adequate visualization knowledge, we create a dataset consisting of abstract utterances and charts and improve model performance through fine-tuning. We further design an interactive interface for ChartGPT that allows users to check and modify the intermediate outputs of each step. The effectiveness of the proposed system is evaluated through quantitative evaluations and a user study.

Efficient and accurate detection of anomalies in photovoltaic (PV) strings is essential for ensuring the normal operation of PV power stations. Most existing studies focus on developing automated anomaly detection models based on temporal abnormalities in PV strings. However, since analyzing anomalies often requires domain knowledge, existing automated methods have significant limitations in assisting experts to understand the causes and impact of these anomalies. In close collaboration with domain experts, this work has summarized the specific user requirements for PV string anomaly detection and designed PVeSight, an interactive visual analysis system to help experts discover and analyze anomalies in PV strings. We use dimensionality reduction techniques to generate string pattern map. These maps are used for anomaly detection, classifying anomalies, comparative analysis between strings, and hierarchical analysis under inverters and combiner boxes. This helps experts trace the causes of anomalies and acquire valuable insights into anomalous PV strings. Through case studies and expert evaluation, we verified the usability and effectiveness of PVeSight for PV string anomaly detection.

The immense popularity of racket sports has fueled substantial demand in tactical analysis with broadcast videos. However, existing manual methods require laborious annotation, and recent attempts leveraging video perception models are limited to low-level annotations like ball trajectories, overlooking tactics that necessitate an understanding of stroke techniques. State-of-the-art action segmentation models also struggle with technique recognition due to frequent occlusions and motion-induced blurring in racket sports videos. To address these challenges, We propose ViSTec, a Video-based Sports Technique recognition model inspired by human cognition that synergizes sparse visual data with rich contextual insights. Our approach integrates a graph to explicitly model strategic knowledge in stroke sequences and enhance technique recognition with contextual inductive bias. A two-stage action perception model is jointly trained to align with the contextual knowledge in the graph. Experiments demonstrate that our method outperforms existing models by a significant margin. Case studies with experts from the Chinese national table tennis team validate our model's capacity to automate analysis for technical actions and tactical strategies. More details are available at: https://ViSTec2024.github.io/.

Visual analytics (VA) systems have been widely used in various application domains. However, VA systems are complex in design, which imposes a serious problem: although the academic community constantly designs and implements new designs, the designs are difficult to query, understand, and refer to by subsequent designers. To mark a major step forward in tackling this problem, we index VA designs in an expressive and accessible way, transforming the designs into a structured format. We first conducted a workshop study with VA designers to learn user requirements for understanding and retrieving professional designs in VA systems. Thereafter, we came up with an index structure VAID to describe advanced and composited visualization designs with comprehensive labels about their analytical tasks and visual designs. The usefulness of VAID was validated through user studies. Our work opens new perspectives for enhancing the accessibility and reusability of professional visualization designs.

In team sports like basketball, understanding and executing tactics—coordinated plans of movements among players—are crucial yet complex, requiring extensive practice. These tactics require players to develop a keen sense of spatial and situational awareness. Traditional coaching methods, which mainly rely on basketball tactic boards and video instruction, often fail to bridge the gap between theoretical learning and the real-world application of tactics, due to shifts in view perspectives and a lack of direct experience with tactical scenarios. To address this challenge, we introduce VisCourt, a Mixed Reality (MR) tactic training system, in collaboration with a professional basketball team. To set up the MR training environment, we employed semi-automatic methods to simulate realistic 3D tactical scenarios and iteratively designed visual in-situ guidance. This approach enables full-body engagement in interactive training sessions on an actual basketball court and provides immediate feedback, significantly enhancing the learning experience. A user study with athletes and enthusiasts shows the effectiveness and satisfaction with VisCourt in basketball training and offers insights for the design of future SportsXR training systems.

In geo-related fields such as urban informatics, atmospheric science, and geography, large-scale spatial time (ST) series (i.e., geo-referred time series) are collected for monitoring and understanding important spatiotemporal phenomena. ST series visualization is an effective means of understanding the data and reviewing spatiotemporal phenomena, which is a prerequisite for in-depth data analysis. However, visualizing these series is challenging due to their large scales, inherent dynamics, and spatiotemporal nature. In this study, we introduce the notion of patterns of evolution in ST series. Each evolution pattern is characterized by 1) a set of ST series that are close in space and 2) a time period when the trends of these ST series are correlated. We then leverage Storyline techniques by considering an analogy between evolution patterns and sessions, and finally design a novel visualization called GeoChron, which is capable of visualizing large-scale ST series in an evolution pattern-aware and narrative-preserving manner. GeoChron includes a mining framework to extract evolution patterns and two-level visualizations to enhance its visual scalability. We evaluate GeoChron with two case studies, an informal user study, an ablation study, parameter analysis, and running time analysis.

In recent years, with growing health consciousness, at-home workout has become increasingly popular for its convenience and safety. Most people choose to follow video guidance during exercising. However, our preliminary study revealed that fitness-minded people face challenges when watching exercise videos on handheld devices or fixed monitors, such as limited movement comprehension due to static camera angles and insufficient feedback. To address these issues, we reviewed popular workout videos, identified user requirements, and came up with an augmented reality (AR) solution. Following a user-centered iterative design process, we proposed a design space of AR visual cues for workouts and implemented an AR-based application. Specifically, we captured users’ exercise performance with pose-tracking technology and provided feedback via AR visual cues. Two user experiments showed that incorporating AR visual cues could improve movement comprehension and enable users to adjust their movements based on real-time feedback. Finally, we presented several suggestions to inspire future design and apply AR visual cues to sports training.

Composite visualization is a popular design strategy that represents complex datasets by integrating multiple visualizations in a meaningful and aesthetic layout, such as juxtaposition, overlay, and nesting. With this strategy, numerous novel designs have been proposed in visualization publications to accomplish various visual analytic tasks. However, there is a lack of understanding of design patterns of composite visualization, thus failing to provide holistic design space and concrete examples for practical use. In this article, we opted to revisit the composite visualizations in IEEE VIS publications and answered what and how visualizations of different types are composed together. To achieve this, we first constructed a corpus of composite visualizations from the publications and analyzed common practices, such as the pattern distributions and co-occurrence of visualization types. From the analysis, we obtained insights into different design patterns on the utilities and their potential pros and cons. Furthermore, we discussed usage scenarios of our taxonomy and corpus and how future research on visualization composition can be conducted on the basis of this study.

Analytical dashboards are popular in business intelligence to facilitate insight discovery with multiple charts. However, creating an effective dashboard is highly demanding, which requires users to have adequate data analysis background and be familiar with professional tools, such as Power BI. To create a dashboard, users have to configure charts by selecting data columns and exploring different chart combinations to optimize the communication of insights, which is trial-and-error. Recent research has started to use deep learning methods for dashboard generation to lower the burden of visualization creation. However, such efforts are greatly hindered by the lack of large-scale and high-quality datasets of dashboards. In this work, we propose using deep reinforcement learning to generate analytical dashboards that can use well-established visualization knowledge and the estimation capacity of reinforcement learning. Specifically, we use visualization knowledge to construct a training environment and rewards for agents to explore and imitate human exploration behavior with a well-designed agent network. The usefulness of the deep reinforcement learning model is demonstrated through ablation studies and user studies. In conclusion, our work opens up new opportunities to develop effective ML-based visualization recommenders without beforehand training datasets.

Images in visualization publications contain rich information, e.g., novel visualization designs and implicit design patterns of visualizations. A systematic collection of these images can contribute to the community in many aspects, such as literature analysis and automated tasks for visualization. In this paper, we build and make public a dataset, VisImages, which collects 12,267 images with captions from 1,397 papers in IEEE InfoVis and VAST. Built upon a comprehensive visualization taxonomy, the dataset includes 35,096 visualizations and their bounding boxes in the images. We demonstrate the usefulness of VisImages through three use cases: 1) investigating the use of visualizations in the publications with VisImages Explorer, 2) training and benchmarking models for visualization classification, and 3) localizing visualizations in the visual analytics systems automatically.

The last decade has witnessed many visual analytics (VA) systems that make successful applications to wide-ranging domains like urban analytics and explainable AI. However, their research rigor and contributions have been extensively challenged within the visualization community. We come in defence of VA systems by contributing two interview studies for gathering critics and responses to those criticisms. First, we interview 24 researchers to collect criticisms the review comments on their VA work. Through an iterative coding and refinement process, the interview feedback is summarized into a list of 36 common criticisms. Second, we interview 17 researchers to validate our list and collect their responses, thereby discussing implications for defending and improving the scientific values and rigor of VA systems. We highlight that the presented knowledge is deep, extensive, but also imperfect, provocative, and controversial, and thus recommend reading with an inclusive and critical eye. We hope our work can provide thoughts and foundations for conducting VA research and spark discussions to promote the research field forward more rigorously and vibrantly.

In a basketball play, players who are not in possession of the ball (i.e., off-ball players) can still effectively contribute to the team's offense, such as making a sudden move to create scoring opportunities. Analyzing the movements of off-ball players can thus facilitate the development of effective strategies for coaches. However, common basketball statistics (e.g., points and assists) primarily focus on what happens around the ball and are mostly result-oriented, making it challenging to objectively assess and fully understand the contributions of off-ball movements. To address these challenges, we collaborate closely with domain experts and summarize the multi-level requirements for off-ball movement analysis in basketball. We first establish an assessment model to quantitatively evaluate the offensive contribution of an off-ball movement considering both the position of players and the team cooperation. Based on the model, we design and develop a visual analytics system called OBTracker to support the multifaceted analysis of off-ball movements. OBTracker enables users to identify the frequency and effectiveness of off-ball movement patterns and learn the performance of different off-ball players. A tailored visualization based on the Voronoi diagram is proposed to help users interpret the contribution of off-ball movements from a temporal perspective. We conduct two case studies based on the tracking data from NBA games and demonstrate the effectiveness and usability of OBTracker through expert feedback.

Glyph-based visualization achieves an impressive graphic design when associated with comprehensive visual metaphors, which help audiences effectively grasp the conveyed information through revealing data semantics. However, creating such metaphoric glyph-based visualization (MGV) is not an easy task, as it requires not only a deep understanding of data but also professional design skills. This paper proposes MetaGlyph, an automatic system for generating MGVs from a spreadsheet. To develop MetaGlyph, we first conduct a qualitative analysis to understand the design of current MGVs from the perspectives of metaphor embodiment and glyph design. Based on the results, we introduce a novel framework for generating MGVs by metaphoric image selection and an MGV construction. Specifically, MetaGlyph automatically selects metaphors with corresponding images from online resources based on the input data semantics. We then integrate a Monte Carlo tree search algorithm that explores the design of an MGV by associating visual elements with data dimensions given the data importance, semantic relevance, and glyph non-overlap. The system also provides editing feedback that allows users to customize the MGVs according to their design preferences. We demonstrate the use of MetaGlyph through a set of examples, one usage scenario, and validate its effectiveness through a series of expert interviews.

The popularity of racket sports (e.g., tennis and table tennis) leads to high demands for data analysis, such as notational analysis, on player performance. While sports videos offer many benefits for such analysis, retrieving accurate information from sports videos could be challenging. In this paper, we propose EventAnchor, a data analysis framework to facilitate interactive annotation of racket sports video with the support of computer vision algorithms. Our approach uses machine learning models in computer vision to help users acquire essential events from videos (e.g., serve, the ball bouncing on the court) and offers users a set of interactive tools for data annotation. An evaluation study on a table tennis annotation system built on this framework shows significant improvement of user performances in simple annotation tasks on objects of interest and complex annotation tasks requiring domain knowledge.

Stroke evaluation is critical for coaches to evaluate players' performance in table tennis matches. However, current methods highly demand proficient knowledge in table tennis and are time-consuming. We collaborate with the Chinese national table tennis team and propose Tac-Valuer, an automatic stroke evaluation framework for analysts in table tennis teams. In particular, to integrate analysts' knowledge into the machine learning model, we employ the latest effective framework named abductive learning, showing promising performance. Based on abductive learning, Tac-Valuer combines the state-of-the-art computer vision algorithms to extract and embed stroke features for evaluation. We evaluate the design choices of the approach and present Tac-Valuer's usability through use cases that analyze the performance of the top table tennis players in world-class events.

In soccer, passing is the most frequent interaction between players and plays a significant role in creating scoring chances. Experts are interested in analyzing players' passing behavior to learn passing tactics, i.e., how players build up an attack with passing. Various approaches have been proposed to facilitate the analysis of passing tactics. However, the dynamic changes of a team's employed tactics over a match have not been comprehensively investigated. To address the problem, we closely collaborate with domain experts and characterize requirements to analyze the dynamic changes of a team's passing tactics. To characterize the passing tactic employed for each attack, we propose a topic-based approach that provides a high-level abstraction of complex passing behaviors. Based on the model, we propose a glyph-based design to reveal the multi-variate information of passing tactics within different phases of attacks, including player identity, spatial context, and formation. We further design and develop PassVizor, a visual analytics system, to support the comprehensive analysis of passing dynamics. With the system, users can detect the changing patterns of passing tactics and examine the detailed passing process for evaluating passing tactics. We invite experts to conduct analysis with PassVizor and demonstrate the usability of the system through an expert interview.

Simulative analysis in competitive sports can provide prospective insights, which can help improve the performance of players in future matches. However, adequately simulating the complex competition process and effectively explaining the simulation result to domain experts are typically challenging. This work presents a design study to address these challenges in table tennis. We propose a well-established hybrid second-order Markov chain model to characterize and simulate the competition process in table tennis. Compared with existing methods, our approach is the first to support the effective simulation of tactics, which represent high-level competition strategies in table tennis. Furthermore, we introduce a visual analytics system called Tac-Simur based on the proposed model for simulative visual analytics. Tac-Simur enables users to easily navigate different players and their tactics based on their respective performance in matches to identify the player and the tactics of interest for further analysis. Then, users can utilize the system to interactively explore diverse simulation tasks and visually explain the simulation results. The effectiveness and usefulness of this work are demonstrated by two case studies, in which domain experts utilize Tac-Simur to find interesting and valuable insights. The domain experts also provide positive feedback on the usability of Tac-Simur. Our work can be extended to other similar sports such as tennis and badminton.

Regarded as a high-level tactic in soccer, a team formation assigns players different tasks and indicates their active regions on the pitch, thereby influencing the team performance significantly. Analysis of formations in soccer has become particularly indispensable for soccer analysts. However, formations of a team are intrinsically time-varying and contain inherent spatial information. The spatio-temporal nature of formations and other characteristics of soccer data, such as multivariate features, make analysis of formations in soccer a challenging problem. In this study, we closely worked with domain experts to characterize domain problems of formation analysis in soccer and formulated several design goals. We design a novel spatio-temporal visual representation of changes in team formation, allowing analysts to visually analyze the evolution of formations and track the spatial flow of players within formations over time. Based on the new design, we further design and develop ForVizor, a visual analytics system, which empowers users to track the spatio-temporal changes in formation and understand how and why such changes occur. With ForVizor, domain experts conduct formation analysis of two games. Analysis results with insights and useful feedback are summarized in two case studies.