Emotionally Aware Moderation: The Potential of Emotion Monitoring in Shaping Healthier Social Media ConversationsSocial media platforms increasingly employ proactive moderation techniques, such as detecting and curbing toxic and uncivil comments, to prevent the spread of harmful content. Despite these efforts, such approaches are often criticized for creating a climate of censorship and failing to address the underlying causes of uncivil behavior. Our work makes both theoretical and practical contributions by proposing and evaluating two types of emotion monitoring dashboards to users' emotional awareness and mitigate hate speech. In a study involving 211 participants, we evaluate the effects of the two mechanisms on user commenting behavior and emotional experiences. The results reveal that these interventions effectively increase users' awareness of their emotional states and reduce hate speech. However, our findings also indicate potential unintended effects, including increased expression of negative emotions (Angry, Fear, and Sad) when discussing sensitive issues. These insights provide a basis for further research on integrating proactive emotion regulation tools into social media platforms to foster healthier digital interactions.2025XSXiaotian Su et al.Toxic and Anti-Social BehaviorCSCW
Preference-Guided Multi-Objective UI Adaptation3D Mixed Reality interfaces have nearly unlimited space for layout placement, making automatic UI adaptation crucial for enhancing the user experience. Such adaptation is often formulated as a multi-objective optimization (MOO) problem, where multiple, potentially conflicting design objectives must be balanced. However, selecting a final layout is challenging since MOO typically yields a set of trade-offs along a Pareto frontier. Prior approaches often required users to manually explore and evaluate these trade-offs, a time-consuming process that disrupts the fluidity of interaction. To eliminate this manual and laborous step, we propose a novel optimization approach that efficiently determines user preferences from a minimal number of UI element adjustments. These determined rankings are translated into priority levels, which then drive our priority-based MOO algorithm. By focusing the search on user-preferred solutions, our method not only identifies UIs that are more aligned with user preferences, but also automatically selects the final design from the Pareto frontier; ultimately, it minimizes user effort while ensuring personalized layouts. Our user study in a Mixed Reality setting demonstrates that our preference-guided approach significantly reduces manual adjustments compared to traditional methods, including fully manual design and exhaustive Pareto front searches, while maintaining high user satisfaction. We believe this work opens the door for more efficient MOO by seamlessly incorporating user preferences.2025YSYao Song et al.Mixed Reality WorkspacesHuman-LLM CollaborationUIST
SwitchAR: Perceptual Manipulations in Augmented RealityPerceptual manipulations (PMs) like redirected walking (RDW) are frequently applied in Virtual Reality (VR) to overcome technological limitations. These PMs manipulate the user’s visual perceptions (e.g. through rotational gains), which is currently challenging in Augmented Reality (AR). We propose SwitchAR, a PM for video pass-through AR leveraging change and inattentional blindness to imperceptibly switch between the camera stream of the real environment and a 3D reconstruction. This enables perceptual manipulations in what users still perceive as AR. We present our pipeline consisting of (1) Reconstruction, (2) Switch (AR -> VR), (3) PM and (4) Switch (VR -> AR), and discuss its challenges and our solutions. In a user study (n=20), we found that no participant noticed the switch and only one the PM. Additionally, despite revealing that a manipulation happened, participants could not detect the switch in a consecutive run. SwitchAR is a fundamental basis enabling AR PMs.2025JWJonas Wombacher et al.AR Navigation & Context AwarenessImmersion & Presence ResearchUIST
DxHF: Providing High-Quality Human Feedback for LLM Alignment with Interactive DecompositionHuman preferences are widely used to align large language models (LLMs) through methods such as reinforcement learning from human feedback (RLHF). However, the current user interfaces require annotators to compare text paragraphs, which is cognitively challenging when the texts are long or unfamiliar. This paper contributes by studying the decomposition principle as an approach to improving the quality of human feedback for LLM alignment. This approach breaks down the text into individual claims instead of directly comparing two long-form text responses. Based on the principle, we build a novel user interface DxHF. It enhances the comparison process by showing decomposed claims, visually encoding the relevance of claims to the conversation and linking similar claims. This allows users to skim through key information and identify differences for better and quicker judgment. Our technical evaluation shows evidence that decomposition generally improves feedback accuracy regarding the ground truth, particularly for users with uncertainty. A crowdsourcing study with 160 participants indicates that using DxHF improves feedback accuracy by an average of 5%, although it increases the average feedback time by 18 seconds. Notably, accuracy is significantly higher in situations where users have less certainty. The finding of the study highlights the potential of HCI as an effective method for improving human-AI alignment.2025DSDanqing Shi et al.Human-LLM CollaborationExplainable AI (XAI)UIST
Efficient Visual Appearance Optimization by Learning from Prior PreferencesAdjusting visual parameters such as brightness and contrast is common in our everyday experiences. Finding the optimal parameter setting is challenging due to the large search space and the lack of an explicit objective function, leaving users to rely solely on their implicit preferences. Prior work has explored Preferential Bayesian Optimization (PBO) to address this challenge, involving users to iteratively select preferred designs from candidate sets. However, PBO often requires many rounds of preference comparisons, making it more suitable for designers than everyday end-users. We propose Meta-PO, a novel method that integrates PBO with meta-learning to improve sample efficiency. Specifically, Meta-PO infers prior users' preferences and stores them as models, which are leveraged to intelligently suggest design candidates for the new users, enabling faster convergence and more personalized results. An experimental evaluation of our method for appearance design tasks on 2D and 3D content showed that participants achieved satisfactory appearance in 5.86 iterations using Meta-PO when participants shared similar goals with a population (e.g., tuning for a ``warm'' look) and in 8 iterations even generalizes across divergent goals (e.g., from ``vintage'', ``warm'', to ``holiday''). Meta-PO makes personalized visual optimization more applicable to end-users through a generalizable, more efficient optimization conditioned on preferences, with the potential to scale interface personalization more broadly.2025ZLZhipeng Li et al.Explainable AI (XAI)AI-Assisted Decision-Making & AutomationUIST
🌳-generAItor: Tree-in-the-loop Text Generation for Language Model Explainability and AdaptationLarge language models (LLMs) are widely deployed in various downstream tasks, e.g., auto-completion, aided writing, or chat-based text generation. However, the considered output candidates of the underlying search algorithm are under-explored and under-explained. We tackle this shortcoming by proposing a tree-in-the-loop approach, where a visual representation of the beam search tree is the central component for analyzing, explaining, and adapting the generated outputs. To support these tasks, we present generAItor, a visual analytics technique, augmenting the central beam search tree with various task-specific widgets, providing targeted visualizations and interaction possibilities. Our approach allows interactions on multiple levels and offers an iterative pipeline that encompasses generating, exploring, and comparing output candidates, as well as fine-tuning the model based on adapted data. Our case study shows that our tool generates new insights in gender bias analysis beyond state-of-the-art template-based methods. Additionally, we demonstrate the applicability of our approach in a qualitative user study. Finally, we quantitatively evaluate the adaptability of the model to few samples, as occurring in text-generation use cases.2025TSThilo Spinner et al.Explainable AI (XAI)Recommender System UXInteractive Data VisualizationIUI
Redefining Affordance via Computational RationalityAffordances, a foundational concept in human-computer interaction and design, have traditionally been explained by direct-perception theories, which assume that individuals perceive action possibilities directly from the environment. However, these theories fall short of explaining how affordances are perceived, learned, refined, or misperceived, and how users choose between multiple affordances in dynamic contexts. This paper introduces a novel affordance theory grounded in Computational Rationality, positing that humans construct internal representations of the world based on bounded sensory inputs. Within these internal models, affordances are inferred through two core mechanisms: feature recognition and hypothetical motion trajectories. Our theory redefines affordance perception as a decision-making process, driven by two components: confidence (the perceived likelihood of successfully executing an action) and predicted utility (the expected value of the outcome). By balancing these factors, individuals make informed decisions about which actions to take. Our theory frames affordances perception as dynamic, continuously learned, and refined through reinforcement and feedback. We validate the theory via thought experiments and demonstrate its applicability across diverse types of affordances (e.g., physical, digital, social). Beyond clarifying and generalizing the understanding of affordances across contexts, our theory serves as a foundation for improving design communication and guiding the development of more adaptive and intuitive systems that evolve with user capabilities.2025YLYi-Chi Liao et al.Explainable AI (XAI)Privacy by Design & User ControlUser Research Methods (Interviews, Surveys, Observation)IUI
A Design Space for Intelligent Dialogue AugmentationThe use of intelligent agents in communication is a growing trend aimed at enhancing the efficiency and quality of interactions. As such, \emph{dialogue augmentation systems}---text processing systems that interactively enhance ongoing written or spoken communication---are gaining significant popularity across domains. While technical limitations had previously inhibited their real-time usage for effective communication augmentation, recent developments in language processing have improved their capabilities to contribute to dialogue as intelligent, emancipated, and proactive agents. While other works on dialogue augmentation focus on evaluating design considerations for specific applications of these systems, we lack a unified understanding of the broader design principles that apply to dialogue more generally. Through a literature review and mixed-methods analysis of 78 existing systems, we iteratively define a comprehensive design space for intelligent dialogue augmentation systems. To further ground our analysis, we interweave Clark's models of dialogue with concepts in human-AI collaboration and discuss trends in the evolving role of dialogue augmentation systems along five dimensions---dialogue context, augmentation context, task, interaction, and model. Based on the identified trends, we discuss concrete challenges for broader adoption, highlighting the need to design \emph{trusted}, \emph{seamless}, \emph{timely}, and \emph{accessible} augmentations. The design space contributes as a mechanism for researchers to facilitate defining design choices during development, situate their systems in the current landscape of works, and understand opportunities for future research.2025RCRobin Shing Moon Chan et al.Conversational ChatbotsAgent Personality & AnthropomorphismHuman-LLM CollaborationIUI
InteRecon: Towards Reconstructing Interactivity of Personal Memorable Items in Mixed RealityDigital capturing of memorable personal items is a key way to archive personal memories. Although current digitization methods (e.g., photos, videos, 3D scanning) can replicate the physical appearance of an item, they often cannot preserve its real-world interactivity. We present Interactive Digital Item (IDI), a concept of reconstructing both the physical appearance and, more importantly, the interactivity of an item. We first conducted a formative study to understand users' expectations of IDI, identifying key physical interactivity features, including geometry, interfaces, and embedded content of items. Informed by these findings, we developed InteRecon, an AR prototype enabling personal reconstruction functions for IDI creation. An exploratory study was conducted to assess the feasibility of using InteRecon and explore the potential of IDI to enrich personal memory archives. Results show that InteRecon is feasible for IDI creation, and the concept of IDI brings new opportunities for augmenting personal memory archives.2025ZLZisu Li et al.The Hong Kong University of Science and Technology, IIP (Computational Media and Arts); MIT CSAILInteractive Narrative & Immersive StorytellingCHI
Beyond Deterrence: A Systematic Review of the Role of Autonomous Motivation in Organizational Security Behavior StudiesWhat drives employees to ensure security when handling information assets in organizations? There is growing interest from the security behavior community in how autonomous motivators shape employees’ security-related behaviors. To reconcile the scattered viewpoints on autonomous motivation and synthesize findings from studies utilizing various theoretical frameworks, we systematically reviewed relevant publications. We present a preregistered literature review that investigated (a) what forms of autonomous motivation have been examined in organizational security contexts, (b) which behaviors/behavioral intentions are related to autonomous motivators, and (c) how autonomous motivation affects employees’ security behaviors. Based on an initial set of 432 papers, filtered down to 45 studies, we identified 17 unique autonomous motivators and three types of related security behaviors. This review not only develops a refined taxonomy of autonomous motivation related to security behaviors but also charts a path forward for future research on autonomous motivation in human-centered security.2025XCXiaowei Chen et al.University of Luxembourg, Institute for Advanced StudiesCybersecurity Training & AwarenessCHI
Sketch2Terrain: AI-Driven Real-Time Terrain Sketch Mapping in Augmented RealitySketch mapping is an effective technique to externalize and communicate spatial information. However, it has been limited to 2D mediums, making it difficult to represent 3D information, particularly for terrains with elevation changes. We present Sketch2Terrain, an intuitive generative-3D-sketch-mapping system combining freehand sketching with generative Artificial Intelligence that radically changes sketch map creation and representation using Augmented Reality. Sketch2Terrain empowers non-experts to create unambiguous sketch maps of natural environments and provides a homogeneous interface for researchers to collect data and conduct experiments. A between-subject study (N=36) revealed that generative-3D-sketch-mapping improved efficiency by 38.4%, terrain-topology accuracy by 12.5%, and landmark accuracy by up to 12.1%, with only a 4.7% trade-off in terrain-elevation accuracy compared to freehand 3D-sketch-mapping. Additionally, generative-3D-sketch-mapping reduced perceived strain by 60.5% and stress by 39.5% over 2D-sketch-mapping. These findings underscore potential applications of generative-3D-sketch-mapping for in-depth understanding and accurate representation of vertically complex environments. The implementation is publicly available.2025TXTianyi Xiao et al.Institute of Cartography and Geoinformation, ETH ZurichAR Navigation & Context AwarenessGenerative AI (Text, Image, Music, Video)Geospatial & Map VisualizationCHI
Moving Beyond the Simulator: Interaction-Based Drunk Driving Detection in a Real Vehicle Using Driver Monitoring Cameras and Real-Time Vehicle DataAlcohol consumption poses a significant public health challenge, presenting serious risks to individual health and contributing to over 700 daily road fatalities worldwide. Digital interventions can play a crucial role in reducing these risks. However, reliable drunk driving detection systems are vital to effectively deliver these interventions. To develop and evaluate such a system, we conducted an interventional study on a test track to collect real vehicle data from 54 participants. Our system reliably identifies non-sober driving with an area under the receiver operating characteristic curve (AUROC) of 0.84 ± 0.11 and driving above the WHO-recommended blood alcohol concentration limit of 0.05 g/dL with an AUROC of 0.80 ± 0.10. Our models rely on well-known physiological drunk driving patterns. To the best of our knowledge, we are the first to (1) rigorously evaluate the potential of (2) driver monitoring cameras and real-time vehicle data for detecting drunk driving in a (3) real vehicle.2025RDRobin Deuber et al.ETH ZürichTeleoperated DrivingHuman Pose & Activity RecognitionCHI
Next-Generation Navigation: Evaluating the Impact of Augmented Reality on Situation Awareness in General Aviation CockpitsFlights in general aviation require pilots to navigate using 2D maps, which splits their attention between the cockpit and the outside environment, reducing situation awareness. Augmented reality (AR) can bridge the gap between the inside and outside world, and thus can resolve the issue of attention switches. In a mixed methods simulator study with 19 pilots, we tested an AR application that integrated invisible and hard-to-see aeronautical data and navigation features with the visible world. Results show that the AR tool enhances and accelerates orientation, and can result in flight trajectories being more accurate with AR than without AR. Situation awareness, measured with a subjective self-rating, was not increased with AR support. Participants voiced concerns about AR content occluding outside features, while positive feedback included use cases in unfamiliar areas and in low visibility, as well as highlighting of hazards.2025ASAdrian Sarbach et al.ETH Zurich, Institute of Cartography and GeoinformationAR Navigation & Context AwarenessContext-Aware ComputingCHI
Finding Needles in Document Haystacks: Augmenting Serendipitous Claim Retrieval WorkflowsPreliminary exploration of vast text corpora for generating and validating hypotheses, typical in academic inquiry, requires flexible navigation and rapid validation of claims. Navigating the corpus by titles, summaries, and abstracts might neglect information, whereas identifying the relevant context-specific claims through in-depth reading is unfeasible with rapidly increasing publication numbers. Our paper identifies three typical user pathways for hypothesis exploration and operationalizes sentence-based retrieval combined with effective contextualization and provenance tracking in a unified workflow. We contribute an interface that augments the previously laborious tasks of claim identification and consistency checking using NLP techniques while balancing user control and serendipity. Use cases, expert interviews, and a user study with 10 participants demonstrate how the proposed workflow enables users to traverse literature corpora in novel and efficient ways. For the evaluation, we instantiate the tool within two independent domains, providing novel insights into the analysis of political discourse and medical research.2025MDMoritz Dück et al.ETH ZurichHuman-LLM CollaborationAI-Assisted Decision-Making & AutomationInteractive Data VisualizationCHI
Do It For Me vs. Do It With Me: Investigating User Perceptions of Different Paradigms of Automation in Copilots for Feature-Rich SoftwareLarge Language Model (LLM)-based in-application assistants, or copilots, can automate software tasks, but users often prefer learning by doing, raising questions about the optimal level of automation for an effective user experience. We investigated two automation paradigms by designing and implementing a fully automated copilot (AutoCopilot) and a semi-automated copilot (GuidedCopilot) that automates trivial steps while offering step-by-step visual guidance. In a user study (N=20) across data analysis and visual design tasks, GuidedCopilot outperformed AutoCopilot in user control, software utility, and learnability, especially for exploratory and creative tasks, while AutoCopilot saved time for simpler visual tasks. A follow-up design exploration (N=10) enhanced GuidedCopilot with task-and state-aware features, including in-context preview clips and adaptive instructions. Our findings highlight the critical role of user control and tailored guidance in designing the next generation of copilots that enhance productivity, support diverse skill levels, and foster deeper software engagement.2025AKAnjali Khurana et al.Simon Fraser University, Computing ScienceHuman-LLM CollaborationAI-Assisted Decision-Making & AutomationCHI
3HANDS Dataset: Learning from Humans for Generating Naturalistic Handovers with Supernumerary Robotic LimbsSupernumerary robotic limbs are robotic structures integrated closely with the user's body, which augment human physical capabilities and necessitate seamless, naturalistic human-machine interaction. For effective assistance in physical tasks, enabling SRLs to hand over objects to humans is crucial. Yet, designing heuristic-based policies for robots is time-consuming, difficult to generalize across tasks, and results in less human-like motion. When trained with proper datasets, generative models are powerful alternatives for creating naturalistic handover motions. We introduce 3HANDS, a novel dataset of object handover interactions between a participant performing a daily activity and another participant enacting a hip-mounted SRL in a naturalistic manner. 3HANDS captures the unique characteristics of SRL interactions: operating in intimate personal space with asymmetric object origins, implicit motion synchronization, and the user’s engagement in a primary task during the handover. To demonstrate the effectiveness of our dataset, we present three models: one that generates naturalistic handover trajectories, another that determines the appropriate handover endpoints, and a third that predicts the moment to initiate a handover. In a user study (N=10), we compare the handover interaction performed with our method compared to a baseline. The findings show that our method was perceived as significantly more natural, less physically demanding, and more comfortable.2025AAArtin Saberpour Abadian et al.Saarland University, Saarland Informatics CampusTeleoperated DrivingHuman-Robot Collaboration (HRC)CHI
Stop the Clock - Counteracting Bias Exploited by Attackers through an Interactive Augmented Reality Phishing TrainingPhishing attacks become increasingly sophisticated in targeting humans and exploiting cognitive biases, e.g., through inducing authority or urgency. Previous approaches to user training focused on URL warnings, textual, or click-based training, yielding mixed results. For more interactive training, uncoupled from users’ screens, we explore the potential of Augmented Reality (AR) technologies to enhance phishing detection. Through visual representations of biases that attackers typically exploit and gesture-based interactions with them, the training aims to enable users to counteract cognitive biases by increasing awareness and suspicion. In a laboratory study with N=117 users, we evaluated phishing detection rates, user interaction with, and feedback on the AR-based training in comparison with a click-based variant and a control condition. Our results show that interactive phishing training addressing cognitive biases increased detection rates by 33% and that interactive elements were well perceived. AR technologies further enhance the training.2025LSLorin Schöni et al.ETH Zurich, Security, Privacy & SocietySocial & Collaborative VRCybersecurity Training & AwarenessContext-Aware ComputingCHI
It's a Match - Enhancing the Fit between Users and Phishing Training through PersonalisationEffective training is essential for enhancing users' ability to detect phishing attempts. Personalised training offers huge potential to more closely align training content with individuals' needs and skill levels. In an online study, we assigned N=342 participants to personalised training or a random training variant to compare their effectiveness. The personalisation was based on a phishing proficiency score calculated from factors such as detection ability, knowledge, and security attitude. After training, the participants demonstrated greater proficiency, with an increased ability to detect phishing emails and higher security attitudes. These effects were most pronounced in the personalised condition, demonstrating the potential of personalisation to improve training outcomes. Overall, personalised training levelled the playing field, efficiently bringing all groups, regardless of their initial proficiency, to a comparable and desired post-training phishing proficiency level. Finally, we derived recommendations for designing personalised phishing training content and assigning users to suitable training programmes.2025LSLorin Schöni et al.ETH Zurich, Security, Privacy & SocietyExplainable AI (XAI)Cybersecurity Training & AwarenessCHI
"It's impressive, but in practice ...": Experiencing a Realistic Digital Transformation in and Beyond the ClassroomSerious games, particularly board games, have long been employed in production management education to teach various concepts. While they have demonstrated educational effectiveness, their integration with emerging Industry 4.0 technologies remains limited. Furthermore, there is a lack of empirical research on how industry practitioners apply these digitization technologies in the workplace. To bridge this gap, we designed a course that integrates digital technologies into a traditional board game. We conducted two studies to evaluate both knowledge gains within the classroom and knowledge transfer back into the manufacturing industry. Our results show an improved understanding of the synergies between production management principles and Industry 4.0 technologies, as well as the real-world challenges students face when attempting to transfer this knowledge. Our work contributes pedagogical and practical perspectives on how technology-enhanced serious games can extend learning in and beyond the classroom.2025XZXiaoyu Zhang et al.City University of Hong Kong, School of Creative MediaSerious & Functional GamesSTEM Education & Science CommunicationMental Health Apps & Online Support CommunitiesCHI
Fear, Fun or None: A Qualitative Quest Towards Unlocking Cybersecurity AttitudesEmployees, once seen as the weakest link in organizational cybersecurity, are now recognized as crucial defenders against malicious attacks. Thus, understanding employee attitudes towards cybersecurity, a major factor driving security behavior, is essential for protecting organizations. Using semi-structured interviews and focus groups, this study holistically explores attitudes toward cybersecurity, its influencing factors, and the employees’ needs for fostering positive attitudes. The study offers in-depth insights into affective, cognitive, and behavioral components of attitudes, ranging from annoyance and fear to appreciation for cybersecurity measures. Influencing key factors include (in)direct cybersecurity experiences and individual perceptions - both highlighting social influences. For developing positive attitudes, employees express needs related to the company's social and cultural framework, communication styles, educational contents and formats. The study contributes to developing effective security strategies that address the individual, social, and organizational factors that shape cybersecurity attitudes, ultimately promoting a stronger organizational security.2025APAlexandra von Preuschen et al.Justus Liebig University GiessenCybersecurity Training & AwarenessOnline Harassment & Counter-ToolsCHI