While the prevalence of hybrid meetings brought tremendous economic and social benefits, it is important to identify and address the unique challenges of cross-space interactions between in-person and remote attendees. We conducted 30 observations and 25 in-depth interviews to investigate cross-space interactions. We found that in-person and remote attendees were able to transcend the segmented space and spontaneously interact, especially when the technology was set up properly; however, remote participants sometimes had to fight for the conversational floor. We found that interactions were sometimes affected by where the conversational power resided. The medium that has more attendees holds the power of speaking and influences behaviors of attendees joining from the other medium. We contribute to the literature by providing new insights that are specific to the context when hybrid meetings are more large scale.

Abstract The rise of Large Language Models (LLMs) and generative visual analytics systems has transformed data‐driven insights, yet significant challenges persist in accurately interpreting users analytical and interaction intents. While language inputs offer flexibility, they often lack precision, making the expression of complex intents inefficient, error‐prone, and time‐intensive. To address these limitations, we investigate the design space of multimodal interactions for generative visual analytics through a literature review and pilot brainstorming sessions. Building on these insights, we introduce a highly extensible workflow that integrates multiple LLM agents for intent inference and visualization generation. We develop InterChat, a generative visual analytics system that combines direct manipulation of visual elements with natural language inputs. This integration enables precise intent communication and supports progressive, visually driven exploratory data analyses. By employing effective prompt engineering, and contextual interaction linking, alongside intuitive visualization and interaction designs, InterChat bridges the gap between user interactions and LLM‐driven visualizations, enhancing both interpretability and usability. Extensive evaluations, including two usage scenarios, a user study, and expert feedback, demonstrate the effectiveness of InterChat. Results show significant improvements in the accuracy and efficiency of handling complex visual analytics tasks, highlighting the potential of multimodal interactions to redefine user engagement and analytical depth in generative visual analytics.

Abstract Real‐world machine learning models require rigorous evaluation before deployment, especially in safety‐critical domains like autonomous driving and surveillance. The evaluation of machine learning models often focuses on data slices, which are subsets of the data that share a set of characteristics. Data slice finding automatically identifies conditions or data subgroups where models underperform, aiding developers in mitigating performance issues. Despite its popularity and effectiveness, data slicing for vision model validation faces several challenges. First, data slicing often needs additional image metadata or visual concepts, and falls short in certain computer vision tasks, such as object detection. Second, understanding data slices is a labor‐intensive and mentally demanding process that heavily relies on the expert's domain knowledge. Third, data slicing lacks a human‐in‐the‐loop solution that allows experts to form hypothesis and test them interactively. To overcome these limitations and better support the machine learning operations lifecycle, we introduce VISLIX, a novel visual analytics framework that employs state‐of‐the‐art foundation models to help domain experts analyze slices in computer vision models. Our approach does not require image metadata or visual concepts, automatically generates natural language insights, and allows users to test data slice hypothesis interactively. We evaluate VISLIX with an expert study and three use cases, that demonstrate the effectiveness of our tool in providing comprehensive insights for validating object detection models.

Mindfulness practice has many mental and physical well-being benefits. With the increased popularity of live stream technologies and the impact of COVID-19, many people have turned to live stream tools to participate in online meditation sessions. To better understand the practices, challenges, and opportunities in live-stream meditation, we conducted a three-month autoethnographic study, during which two researchers participated in live-stream meditation sessions as the audience. Then we conducted a follow-up semi-structured interview study with 10 experienced live meditation teachers who use different live-stream tools. We found that live meditation, although having a weaker social presence than in-person meditation, facilitates attendees in establishing a practice routine and connecting with other meditators. Teachers use live streams to deliver the meditation practice to the world which also enhances their practice and brand building. We identified the challenges of using live-stream tools for meditation from the perspectives of both audiences and teachers, and provided design recommendations to better utilize live meditation as a resource for mental wellbeing.

Open-world object detection (OWOD) is an emerging computer vision problem that involves not only the identification of predefined object classes, like what general object detectors do, but also detects new unknown objects simultaneously. Recently, several end-to-end deep learning models have been proposed to address the OWOD problem. However, these approaches face several challenges: a) significant changes in both network architecture and training procedure are required; b) they are trained from scratch, which can not leverage existing pre-trained general detectors; c) costly annotations for all unknown classes are needed. To overcome these challenges, we present a visual analytic framework called OW-Adapter. It acts as an adaptor to enable pre-trained general object detectors to handle the OWOD problem. Specifically, OW-Adapter is designed to identify, summarize, and annotate unknown examples with minimal human effort. Moreover, we introduce a lightweight classifier to learn newly annotated unknown classes and plug the classifier into pre-trained general detectors to detect unknown objects. We demonstrate the effectiveness of our framework through two case studies of different domains, including common object recognition and autonomous driving. The studies show that a simple yet powerful adaptor can extend the capability of pre-trained general detectors to detect unknown objects and improve the performance on known classes simultaneously.

The COVID-19 pandemic greatly changed many people's daily lives. Because of the temporary closure of gyms and fitness centers, many people started to take group fitness classes online. The abrupt transition to synchronous online exercise brought lots of uncertainty and challenges to fitness instructors and participants. This study aims to understand how people teach and participate in live stream group fitness classes, the challenges they have encountered, and their practices to address these challenges. We conducted semi-structured interviews with 11 instructors and 14 participants who had experience teaching or taking live stream group fitness classes on Zoom during the pandemic. Our interviews showed that instructors saw teaching online as a new, more cognitively demanding experience than teaching in-person classes; they found limited resources to create a professional workout space, experienced stress and anxiety, had limited information and approaches to give participants feedback, and faced difficulties with rapport building. Participants found it hard to give real-time reactions while exercising far away from their laptops; they receive less customized instructions, and they have difficulty engaging in private chats. Despite the challenges, we envision CMC in exercise is an opportunity for new design and interactions. We propose design recommendations, including using smartwatches to give real-time reactions and smart garments to generate tactile feedback. Health and biometric data can be shared with instructors and other participants during exercise for safety purposes and to create social connections.

Alerting customers on suspected online-payment fraud and persuade them to terminate transactions is increasingly requested with the rapid growth of digital finance worldwide. We explored the feasibility of using a conversational agent (CA) to fulfill this request. Shing, a voice-based CA, proactively initializes and repairs the conversation with empathetical communication skills in order to alert customers when a suspected online-payment fraud is detected, collects important information for fraud scrutiny and persuades customers to terminate the transaction once the fraud is confirmed. We evaluated our system by comparing it with a rule-based CA with regards to customer response and perceptions in a real-world context where our systems took 144,795 phone calls in total in which 83,019 (57.3%) natural breakdowns happened. Results showed that more customers stopped risky transactions after conversing with Shing. They seemed more willing to converse with Shing for more dialogue turns and provide transaction details. Our work presents practical implications for the design of proactive CA.

In this article, a manifold learning algorithm based on straight-like geodesics and local coordinates is proposed, called SGLC-ML for short. The contribution and innovation of SGLC-ML lie in that; first, SGLC-ML divides the manifold data into a number of straight-like geodesics, instead of a number of local areas like many manifold learning algorithms do. Figuratively speaking, SGLC-ML covers manifold data set with a sparse net woven with threads (straight-like geodesics), while other manifold learning algorithms with a tight roof made of titles (local areas). Second, SGLC-ML maps all straight-like geodesics into straight lines of a low-dimensional Euclidean space. All these straight lines start from the same point and extend along the same coordinate axis. These straight lines are exactly the local coordinates of straight-like geodesics as described in the mathematical definition of the manifold. With the help of local coordinates, dimensionality reduction can be divided into two relatively simple processes: calculation and alignment of local coordinates. However, many manifold learning algorithms seem to ignore the advantages of local coordinates. The experimental results between SGLC-ML and other state-of-the-art algorithms are presented to verify the good performance of SGLC-ML.