KnowledgeVIS: Interpreting Language Models by Comparing Fill-in-the-Blank Prompts.

Recent growth in the popularity of large language models has led to their increased usage for summarizing, predicting, and generating text, making it vital to help researchers and engineers understand how and why they work. We present KnowledgeVIS, a human-in-the-loop visual analytics system for interpreting language models using fill-in-the-blank sentences as prompts. By comparing predictions between sentences, KnowledgeVIS reveals learned associations that intuitively connect what language models learn during training to natural language tasks downstream, helping users create and test multiple prompt variations, analyze predicted words using a novel semantic clustering technique, and discover insights using interactive visualizations. Collectively, these visualizations help users identify the likelihood and uniqueness of individual predictions, compare sets of predictions between prompts, and summarize patterns and relationships between predictions across all prompts. We demonstrate the capabilities of KnowledgeVIS with feedback from six NLP experts as well as three different use cases: (1) probing biomedical knowledge in two domain-adapted models; and (2) evaluating harmful identity stereotypes and (3) discovering facts and relationships between three general-purpose models.

Preliminary Guidelines For Combining Data Integration and Visual Data Analysis.

Data integration is often performed to consolidate information from multiple disparate data sources during visual data analysis. However, integration operations are usually separate from visual analytics operations such as encode and filter in both interface design and empirical research. We conducted a preliminary user study to investigate whether and how data integration should be incorporated directly into the visual analytics process. We used two interface alternatives featuring contrasting approaches to the data preparation and analysis workflow: manual file-based ex-situ integration as a separate step from visual analytics operations; and automatic UI-based in-situ integration merged with visual analytics operations. Participants were asked to complete specific and free-form tasks with each interface, browsing for patterns, generating insights, and summarizing relationships between attributes distributed across multiple files. Analyzing participants' interactions and feedback, we found both task completion time and total interactions to be similar across interfaces and tasks, as well as unique integration strategies between interfaces and emergent behaviors related to satisficing and cognitive bias. Participants' time spent and interactions emergent strategies revealed that in-situ integration enabled users to spend more time on analysis tasks compared with ex-situ integration. Participants' integration strategies and analytical behaviors revealed differences in interface usage for generating and tracking hypotheses and insights , yet their emergent behaviors suggested that in-situ integration could negatively affect the ability to generate and track hypotheses and insights. With these results, we synthesized preliminary guidelines for designing future visual analytics interfaces that can support integrating attributes throughout an active analysis process.

Lumos: Increasing Awareness of Analytic Behavior during Visual Data Analysis.

Visual data analysis tools provide people with the agency and flexibility to explore data using a variety of interactive functionalities. However, this flexibility may introduce potential consequences in situations where users unknowingly overemphasize or underemphasize specific subsets of the data or attribute space they are analyzing. For example, users may overemphasize specific attributes and/or their values (e.g., Gender is always encoded on the X axis), underemphasize others (e.g., Religion is never encoded), ignore a subset of the data (e.g., older people are filtered out), etc. In response, we present Lumos, a visual data analysis tool that captures and shows the interaction history with data to increase awareness of such analytic behaviors. Using in-situ (at the place of interaction) and ex-situ (in an external view) visualization techniques, Lumos provides real-time feedback to users for them to reflect on their activities. For example, Lumos highlights datapoints that have been previously examined in the same visualization (in-situ) and also overlays them on the underlying data distribution (i.e., baseline distribution) in a separate visualization (ex-situ). Through a user study with 24 participants, we investigate how Lumos helps users' data exploration and decision-making processes. We found that Lumos increases users' awareness of visual data analysis practices in real-time, promoting reflection upon and acknowledgement of their intentions and potentially influencing subsequent interactions.

Left, Right, and Gender: Exploring Interaction Traces to Mitigate Human Biases.

Human biases impact the way people analyze data and make decisions. Recent work has shown that some visualization designs can better support cognitive processes and mitigate cognitive biases (i.e., errors that occur due to the use of mental "shortcuts"). In this work, we explore how visualizing a user's interaction history (i.e., which data points and attributes a user has interacted with) can be used to mitigate potential biases that drive decision making by promoting conscious reflection of one's analysis process. Given an interactive scatterplot-based visualization tool, we showed interaction history in real-time while exploring data (by coloring points in the scatterplot that the user has interacted with), and in a summative format after a decision has been made (by comparing the distribution of user interactions to the underlying distribution of the data). We conducted a series of in-lab experiments and a crowd-sourced experiment to evaluate the effectiveness of interaction history interventions toward mitigating bias. We contextualized this work in a political scenario in which participants were instructed to choose a committee of 10 fictitious politicians to review a recent bill passed in the U.S. state of Georgia banning abortion after 6 weeks, where things like gender bias or political party bias may drive one's analysis process. We demonstrate the generalizability of this approach by evaluating a second decision making scenario related to movies. Our results are inconclusive for the effectiveness of interaction history (henceforth referred to as interaction traces) toward mitigating biased decision making. However, we find some mixed support that interaction traces, particularly in a summative format, can increase awareness of potential unconscious biases.

The Role of Interactive Visualization in Fostering Trust in AI.

The increasing use of artificial intelligence (AI) technologies across application domains has prompted our society to pay closer attention to AI's trustworthiness, fairness, interpretability, and accountability. In order to foster trust in AI, it is important to consider the potential of interactive visualization, and how such visualizations help build trust in AI systems. This manifesto discusses the relevance of interactive visualizations and makes the following four claims: i) trust is not a technical problem, ii) trust is dynamic, iii) visualization cannot address all aspects of trust, and iv) visualization is crucial for human agency in AI.

SafetyLens: Visual Data Analysis of Functional Safety of Vehicles.

Modern automobiles have evolved from just being mechanical machines to having full-fledged electronics systems that enhance vehicle dynamics and driver experience. However, these complex hardware and software systems, if not properly designed, can experience failures that can compromise the safety of the vehicle, its occupants, and the surrounding environment. For example, a system to activate the brakes to avoid a collision saves lives when it functions properly, but could lead to tragic outcomes if the brakes were applied in a way that's inconsistent with the design. Broadly speaking, the analysis performed to minimize such risks falls into a systems engineering domain called Functional Safety. In this paper, we present SafetyLens, a visual data analysis tool to assist engineers and analysts in analyzing automotive Functional Safety datasets. SafetyLens combines techniques including network exploration and visual comparison to help analysts perform domain-specific tasks. This paper presents the design study with domain experts that resulted in the design guidelines, the tool, and user feedback.

CAVA: A Visual Analytics System for Exploratory Columnar Data Augmentation Using Knowledge Graphs.

Most visual analytics systems assume that all foraging for data happens before the analytics process; once analysis begins, the set of data attributes considered is fixed. Such separation of data construction from analysis precludes iteration that can enable foraging informed by the needs that arise in-situ during the analysis. The separation of the foraging loop from the data analysis tasks can limit the pace and scope of analysis. In this paper, we present CAVA, a system that integrates data curation and data augmentation with the traditional data exploration and analysis tasks, enabling information foraging in-situ during analysis. Identifying attributes to add to the dataset is difficult because it requires human knowledge to determine which available attributes will be helpful for the ensuing analytical tasks. CAVA crawls knowledge graphs to provide users with a a broad set of attributes drawn from external data to choose from. Users can then specify complex operations on knowledge graphs to construct additional attributes. CAVA shows how visual analytics can help users forage for attributes by letting users visually explore the set of available data, and by serving as an interface for query construction. It also provides visualizations of the knowledge graph itself to help users understand complex joins such as multi-hop aggregations. We assess the ability of our system to enable users to perform complex data combinations without programming in a user study over two datasets. We then demonstrate the generalizability of CAVA through two additional usage scenarios. The results of the evaluation confirm that CAVA is effective in helping the user perform data foraging that leads to improved analysis outcomes, and offer evidence in support of integrating data augmentation as a part of the visual analytics pipeline.

Geono-Cluster: Interactive Visual Cluster Analysis for Biologists.

Biologists often perform clustering analysis to derive meaningful patterns, relationships, and structures from data instances and attributes. Though clustering plays a pivotal role in biologists' data exploration, it takes non-trivial efforts for biologists to find the best grouping in their data using existing tools. Visual cluster analysis is currently performed either programmatically or through menus and dialogues in many tools, which require parameter adjustments over several steps of trial-and-error. In this article, we introduce Geono-Cluster, a novel visual analysis tool designed to support cluster analysis for biologists who do not have formal data science training. Geono-Cluster enables biologists to apply their domain expertise into clustering results by visually demonstrating how their expected clustering outputs should look like with a small sample of data instances. The system then predicts users' intentions and generates potential clustering results. Our study follows the design study protocol to derive biologists' tasks and requirements, design the system, and evaluate the system with experts on their own dataset. Results of our study with six biologists provide initial evidence that Geono-Cluster enables biologists to create, refine, and evaluate clustering results to effectively analyze their data and gain data-driven insights. At the end, we discuss lessons learned and implications of our study.

ArchiText: Interactive Hierarchical Topic Modeling.

Human-in-the-loop topic modeling allows users to explore and steer the process to produce better quality topics that align with their needs. When integrated into visual analytic systems, many existing automated topic modeling algorithms are given interactive parameters to allow users to tune or adjust them. However, this has limitations when the algorithms cannot be easily adapted to changes, and it is difficult to realize interactivity closely supported by underlying algorithms. Instead, we emphasize the concept of tight integration, which advocates for the need to co-develop interactive algorithms and interactive visual analytic systems in parallel to allow flexibility and scalability. In this article, we describe design goals for efficiently and effectively executing the concept of tight integration among computation, visualization, and interaction for hierarchical topic modeling of text data. We propose computational base operations for interactive tasks to achieve the design goals. To instantiate our concept, we present ArchiText, a prototype system for interactive hierarchical topic modeling, which offers fast, flexible, and algorithmically valid analysis via tight integration. Utilizing interactive hierarchical topic modeling, our technique lets users generate, explore, and flexibly steer hierarchical topics to discover more informed topics and their document memberships.

Putting the "I" in Interaction: Interactive Interfaces Personalized to Individuals.

Interactive data exploration and analysis is an inherently personal process. One's background, experience, interests, cognitive style, personality, and other sociotechnical factors often shape such a process, as well as the provenance of exploring, analyzing, and interpreting data. This Viewpoint posits both what personal information and how such personal information could be taken into account to design more effective visual analytic systems, a valuable and under-explored direction.

Investigating Direct Manipulation of Graphical Encodings as a Method for User Interaction.

We investigate direct manipulation of graphical encodings as a method for interacting with visualizations. There is an increasing interest in developing visualization tools that enable users to perform operations by directly manipulating graphical encodings rather than external widgets such as checkboxes and sliders. Designers of such tools must decide which direct manipulation operations should be supported, and identify how each operation can be invoked. However, we lack empirical guidelines for how people convey their intended operations using direct manipulation of graphical encodings. We address this issue by conducting a qualitative study that examines how participants perform 15 operations using direct manipulation of standard graphical encodings. From this study, we 1) identify a list of strategies people employ to perform each operation, 2) observe commonalities in strategies across operations, and 3) derive implications to help designers leverage direct manipulation of graphical encoding as a method for user interaction.

EmoCo: Visual Analysis of Emotion Coherence in Presentation Videos.

Emotions play a key role in human communication and public presentations. Human emotions are usually expressed through multiple modalities. Therefore, exploring multimodal emotions and their coherence is of great value for understanding emotional expressions in presentations and improving presentation skills. However, manually watching and studying presentation videos is often tedious and time-consuming. There is a lack of tool support to help conduct an efficient and in-depth multi-level analysis. Thus, in this paper, we introduce EmoCo, an interactive visual analytics system to facilitate efficient analysis of emotion coherence across facial, text, and audio modalities in presentation videos. Our visualization system features a channel coherence view and a sentence clustering view that together enable users to obtain a quick overview of emotion coherence and its temporal evolution. In addition, a detail view and word view enable detailed exploration and comparison from the sentence level and word level, respectively. We thoroughly evaluate the proposed system and visualization techniques through two usage scenarios based on TED Talk videos and interviews with two domain experts. The results demonstrate the effectiveness of our system in gaining insights into emotion coherence in presentations.

A Provenance Task Abstraction Framework.

Visual analytics tools integrate provenance recording to externalize analytic processes or user insights. Provenance can be captured on varying levels of detail, and in turn activities can be characterized from different granularities. However, current approaches do not support inferring activities that can only be characterized across multiple levels of provenance. We propose a task abstraction framework that consists of a three stage approach, composed of 1) initializing a provenance task hierarchy, 2) parsing the provenance hierarchy by using an abstraction mapping mechanism, and 3) leveraging the task hierarchy in an analytical tool. Furthermore, we identify implications to accommodate iterative refinement, context, variability, and uncertainty during all stages of the framework. We describe a use case which exemplifies our abstraction framework, demonstrating how context can influence the provenance hierarchy to support analysis. The article concludes with an agenda, raising and discussing challenges that need to be considered for successfully implementing such a framework.

BEAMES: Interactive Multimodel Steering, Selection, and Inspection for Regression Tasks.

Interactive model steering helps people incrementally build machine learning models that are tailored to their domain and task. Existing visual analytic tools allow people to steer a single model (e.g., assignment attribute weights used by a dimension reduction model). However, the choice of model is critical in such situations. What if the model chosen is suboptimal for the task, dataset, or question being asked? What if instead of parameterizing and steering this model, a different model provides a better fit? This paper presents a technique to allow users to inspect and steer multiple machine learning models. The technique steers and samples models from a broader set of learning algorithms and model types. We incorporate this technique into a visual analytic prototype, BEAMES, that allows users to perform regression tasks via multimodel steering. This paper demonstrates the effectiveness of BEAMES via a use case, and discusses broader implications for multimodel steering.

Broadening Intellectual Diversity in Visualization Research Papers.

Promoting a wider range of contribution types can facilitate healthy growth of the visualization community, while increasing the intellectual diversity of visualization research papers. In this paper, we discuss the importance of contribution types and summarize contribution types that can be meaningful in visualization research. We also propose several concrete next steps we can and should take to ensure a successful launch of the contribution types.

Demonstrational Interaction for Data Visualization.

Recently, there has been an increasing trend to extend the demonstrational interaction paradigm to visualization tools. As more analytic operations can be performed by demonstration, new user tasks can be supported. In this paper, we discuss the properties of tasks where the by-demonstration paradigm can be effective and describe the main components needed to implement the demonstrational paradigm in visualization tools.

Task-Based Effectiveness of Basic Visualizations.

Visualizations of tabular data are widely used; understanding their effectiveness in different task and data contexts is fundamental to scaling their impact. However, little is known about how basic tabular data visualizations perform across varying data analysis tasks. In this paper, we report results from a crowdsourced experiment to evaluate the effectiveness of five small scale (5-34 data points) two-dimensional visualization types-Table, Line Chart, Bar Chart, Scatterplot, and Pie Chart-across ten common data analysis tasks using two datasets. We find the effectiveness of these visualization types significantly varies across task, suggesting that visualization design would benefit from considering context-dependent effectiveness. Based on our findings, we derive recommendations on which visualizations to choose based on different tasks. We finally train a decision tree on the data we collected to drive a recommender, showcasing how to effectively engineer experimental user data into practical visualization systems.

A Heuristic Approach to Value-Driven Evaluation of Visualizations.

To interpret data visualizations, people must determine how visual features map onto concepts. For example, to interpret colormaps, people must determine how dimensions of color (e.g., lightness, hue) map onto quantities of a given measure (e.g., brain activity, correlation magnitude). This process is easier when the encoded mappings in the visualization match people's predictions of how visual features will map onto concepts, their inferred mappings. To harness this principle in visualization design, it is necessary to understand what factors determine people's inferred mappings. In this study, we investigated how inferred color-quantity mappings for colormap data visualizations were influenced by the background color. Prior literature presents seemingly conflicting accounts of how the background color affects inferred color-quantity mappings. The present results help resolve those conflicts, demonstrating that sometimes the background has an effect and sometimes it does not, depending on whether the colormap appears to vary in opacity. When there is no apparent variation in opacity, participants infer that darker colors map to larger quantities (dark-is-more bias). As apparent variation in opacity increases, participants become biased toward inferring that more opaque colors map to larger quantities (opaque-is-more bias). These biases work together on light backgrounds and conflict on dark backgrounds. Under such conflicts, the opaque-is-more bias can negate, or even supersede the dark-is-more bias. The results suggest that if a design goal is to produce colormaps that match people's inferred mappings and are robust to changes in background color, it is beneficial to use colormaps that will not appear to vary in opacity on any background color, and to encode larger quantities in darker colors.

Augmenting Visualizations with Interactive Data Facts to Facilitate Interpretation and Communication.

Recently, an increasing number of visualization systems have begun to incorporate natural language generation (NLG) capabilities into their interfaces. NLG-based visualization systems typically leverage a suite of statistical functions to automatically extract key facts about the underlying data and surface them as natural language sentences alongside visualizations. With current systems, users are typically required to read the system-generated sentences and mentally map them back to the accompanying visualization. However, depending on the features of the visualization (e.g., visualization type, data density) and the complexity of the data fact, mentally mapping facts to visualizations can be a challenging task. Furthermore, more than one visualization could be used to illustrate a single data fact. Unfortunately, current tools provide little or no support for users to explore such alternatives. In this paper, we explore how system-generated data facts can be treated as interactive widgets to help users interpret visualizations and communicate their findings. We present Voder, a system that lets users interact with automatically-generated data facts to explore both alternative visualizations to convey a data fact as well as a set of embellishments to highlight a fact within a visualization. Leveraging data facts as interactive widgets, Voder also facilitates data fact-based visualization search. To assess Voder's design and features, we conducted a preliminary user study with 12 participants having varying levels of experience with visualization tools. Participant feedback suggested that interactive data facts aided them in interpreting visualizations. Participants also stated that the suggestions surfaced through the facts helped them explore alternative visualizations and embellishments to communicate individual data facts.

Embedded Merge & Split: Visual Adjustment of Data Grouping.

Data grouping is among the most frequently used operations in data visualization. It is the process through which relevant information is gathered, simplified, and expressed in summary form. Many popular visualization tools support automatic grouping of data (e.g., dividing up a numerical variable into bins). Although grouping plays a pivotal role in supporting data exploration, further adjustment and customization of auto-generated grouping criteria is non-trivial. Such adjustments are currently performed either programmatically or through menus and dialogues which require specific parameter adjustments over several steps. In response, we introduce Embedded Merge & Split (EMS), a new interaction technique for direct adjustment of data grouping criteria. We demonstrate how the EMS technique can be designed to directly manipulate width and position in bar charts and histograms, as a means for adjustment of data grouping criteria. We also offer a set of design guidelines for supporting EMS. Finally, we present the results of two user studies, providing initial evidence that EMS can significantly reduce interaction time compared to WIMP-based technique and was subjectively preferred by participants.