arXiv subjects

Benjamin Tag

Publications and source records attributed to Benjamin Tag.

At least 19 recordsLinked to original sources

arXiv2026

Group Alignment-Induced Sycophancy: A Two-Sided Evaluation of Steerable Pluralistic Alignment

Group alignment adapts a language model to a demographic group to produce responses that reflect the group's opinions, values, and preferences. Sycophancy, a well-documented by-product of alignment, causes the model to over-agree with the user regardless of factual and objective information. However, existing group alignment methods and evaluations focus only on how closely the model matches the group's opinions, overlooking the induced change in sycophantic behaviour. To bridge this gap, we introduce \textbf{G}roup \textbf{A}lignment-induced \textbf{S}ycophancy (GAS) and systematically evaluate alignment across 3 methods, 4 models and 13 demographic groups, on both the intended gain in opinion alignment and the unintended shift in sycophancy. We find that gain and shift are non-uniform across groups: under an identical budget, some groups receive larger gains in opinion alignment than others, and the induced sycophancy shift forms a group-specific profile rather than a single-dimensional change. These results suggest that group alignment should be reported as a two-sided, multi-dimensional profile rather than a single fit score that accounts for per-group differences when adapting LLMs to diverse populations.

cs.CL↗

arXiv2026

Generating Natural and Expressive Robot Gestures through Iterative Reinforcement Learning with Human Feedback using LLMs

Expressive gestures are essential for natural and effective communication, complementing speech when verbal cues alone are insufficient (e.g., pointing). For social robots such as the humanoid Pepper, producing natural and expressive movements is critical for improving human-robot interaction (HRI) and long-term acceptance. However, generating gestures remains challenging due to reliance on expert-authored animations, resulting in rigid behaviors that are impractical for dynamic and diverse environments. Alternatively, machine learning approaches often struggle to capture perceived naturalness, becoming increasingly challenging with more degrees of freedom. Consequently, producing expressive robot gestures requires a system that can adapt to the environment while adhering to social norms and physical constraints. Recent advances in large language models (LLMs) enable dynamic code generation, offering new opportunities for runtime gesture synthesis from natural language. In this paper, we integrate ChatGPT into the humanoid robot Pepper to generate co-speech gestures aligned with conversational output. While this baseline enables flexible gesture generation, the resulting motions are often perceived as stiff and unnatural. To address this limitation, we introduce an iterative reinforcement learning with human feedback (RLHF) system that finetunes gesture generation based on user evaluations, leveraging an iterative user study to compare Pepper's generated gestures. Our results show that RLHF improved the LLM's co-speech generative capabilities, producing more expressive, relevant and fluid movements.

cs.RO↗

arXiv2026

AnyMo: Geometry-Aware Setup-Agnostic Modeling of Human Motion in the Wild

As wearable and mobile devices become increasingly embedded in daily life, they offer a practical way to continuously sense human motion in the wild. But inertial signals are highly dependent on the sensing setup, including body location, mounting position, sensor orientation, device hardware, and sampling protocol. This setup dependence makes it difficult to learn motion representations that transfer across devices and datasets, and limits the broader use of wearable IMUs beyond closed-set recognition. We introduce AnyMo, a geometry-aware framework for setup-agnostic human motion modeling. AnyMo uses physics-grounded IMU simulation over dense body-surface placements to generate diverse and plausible synthetic signals, pre-trains a graph encoder from paired synthetic placement views and masked partial observations, tokenizes multi-position IMU into full-body motion tokens, and aligns these tokens with an LLM for motion-language understanding. We evaluate AnyMo on three complementary tasks: zero-shot activity recognition across 14 unseen downstream datasets, cross-modal retrieval, and wearable IMU motion captioning, where it improves average Accuracy/F1/R@2 by 11.7\%/11.6\%/22.6\% on HAR, increases zero-shot IMU-to-text and text-to-IMU retrieval MRR by 15.9\% and 28.6\%, respectively, and improves zero-shot captioning BERT-F1 by 18.8\%. These results support AnyMo as a generalist model for wearable motion understanding in the wild. Project page: https://baiyuchen.com/project/AnyMo.

cs.CV↗

arXiv2026

An LLM-Assisted Toolkit for Inspectable Multimodal Emotion Data Annotation

Multimodal Emotion Recognition (MER) increasingly depends on fine grained, evidence grounded annotations, yet inspection and label construction are hard to scale when cues are dynamic and misaligned across modalities. We present an LLM-assisted toolkit that supports multimodal emotion data annotation through an inspectable, event centered workflow. The toolkit preprocesses and aligns heterogeneous recordings, visualizes all modalities on an interactive shared timeline, and renders structured signals as video tracks for cross modal consistency checks. It then detects candidate events and packages synchronized keyframes and time windows as event packets with traceable pointers to the source data. Finally, the toolkit integrates an LLM with modality specific tools and prompt templates to draft structured annotations for analyst verification and editing. We demonstrate the workflow on multimodal VR emotion recordings with representative examples.

cs.HC↗

arXiv2026

Understanding the Effects of Interaction on Emotional Experiences in VR

Virtual reality has been effectively used for eliciting emotions, yet most research focuses on the intensity of affective responses rather than on how interaction influences those experiences. To address this gap, we advance a validated VR emotion-elicitation dataset through two key extensions. First, we add a new high-arousal, high-valence scene and validate its effectiveness in a within-subject study (N=24). Second, we incorporate interactive elements into each scene, creating both interactive and non-interactive versions to examine the impact of interaction on emotional responses. We evaluate interaction through a multimodal approach combining subjective ratings and physiological signals to capture both conscious and unconscious affective responses. Our evaluation study (N=84) shows that interaction not only amplifies emotions but modulates them in context, supporting coping in negative scenes and enhancing enjoyment in positive scenes. These findings highlight the potential of scene-tailored interaction for different applications, where regulating emotions is as important as eliciting them.

cs.HC↗

arXiv2026

Robot-Wearable Conversation Hand-off for Navigation

Navigating large and complex indoor environments, such as universities, airports, and hospitals, can be cognitively demanding and requires attention and effort. While mobile applications provide convenient navigation support, they occupy the user's hands and visual attention, limiting natural interaction. In this paper, we explore conversation hand-off as a method for multi-device indoor navigation, where a Conversational Agent (CA) transitions seamlessly from a stationary social robot to a wearable device. We evaluated robot-only, wearable-only, and robot-to-wearable hand-off in a university campus setting using a within-subjects design with N=24 participants. We find that conversation hand-off is experienced as engaging, even though no performance benefits were observed, and most preferred using the wearable-only system. Our findings suggest that the design of such re-embodied assistants should maintain a shared voice and state across embodiments. We demonstrate how conversational hand-offs can bridge cognitive and physical transitions, enriching human interaction with embodied AI.

cs.HC↗

arXiv2026

Human Factors in Immersive Analytics

It has been ten years since the term ''Immersive Analytics'' (IA) was coined and research interest in the topic remains strong. Researchers in this field have produced practical and conceptual knowledge concerning the use of emerging immersive spatial display and interaction technologies for sense-making tasks through a number of papers, surveys, and books. However, a lack of truly physically and psychologically ergonomic techniques, as well as standardized human-centric validation protocols for these, remains a significant barrier to wider acceptance of practical IA systems in ubiquitous applications. Building upon a series of workshops on immersive analytics at various conferences, this workshop aims to explore new approaches and establish standard practices for evaluating immersive analytics systems from a human factors perspective. We will gather immersive analytics researchers and practitioners to look closely at these human factors -- including cognitive and physical functions as well as behaviour and performance -- to see how they inform the design and deployment of immersive analytics techniques and applications and to inform future research.

cs.HC↗

arXiv2026

From Fixed to Flexible: Shaping AI Personality in Context-Sensitive Interaction

Conversational agents are increasingly expected to adapt across contexts and evolve their personalities through interactions, yet most remain static once configured. We present an exploratory study of how user expectations form and evolve when agent personality is made dynamically adjustable. To investigate this, we designed a prototype conversational interface that enabled users to adjust an agent's personality along eight research-grounded dimensions across three task contexts: informational, emotional, and appraisal. We conducted an online mixed-methods study with 60 participants, employing latent profile analysis to characterize personality classes and trajectory analysis to trace evolving patterns of personality adjustment. These approaches revealed distinct personality profiles at initial and final configuration stages, and adjustment trajectories, shaped by context-sensitivity. Participants also valued the autonomy, perceived the agent as more anthropomorphic, and reported greater trust. Our findings highlight the importance of designing conversational agents that adapt alongside their users, advancing more responsive and human-centred AI.

cs.HC↗

arXiv2025

When Ads Become Profiles: Uncovering the Invisible Risk of Web Advertising at Scale with LLMs

Regulatory limits on explicit targeting have not eliminated algorithmic profiling on the Web, as optimisation systems still adapt ad delivery to users' private attributes. The widespread availability of powerful zero-shot multimodal Large Language Models (LLMs) has dramatically lowered the barrier for exploiting these latent signals for adversarial inference. We investigate this emerging societal risk, specifically how adversaries can now exploit these signals to reverse-engineer private attributes from ad exposure alone. We introduce a novel pipeline that leverages LLMs as adversarial inference engines to perform natural language profiling. Applying this method to a longitudinal dataset comprising over 435,000 Facebook ad impressions collected from 891 users, we conducted a large-scale study to assess the feasibility and precision of inferring private attributes from passive online ad observations. Our results demonstrate that off-the-shelf LLMs can accurately reconstruct complex user private attributes, including party preference, employment status, and education level, consistently outperforming strong census-based priors and matching or exceeding human social perception at only a fraction of the cost (223x lower) and time (52x faster) required by humans. Critically, actionable profiling is feasible even within short observation windows, indicating that prolonged tracking is not a prerequisite for a successful attack. These findings provide the first empirical evidence that ad streams serve as a high-fidelity digital footprint, enabling off-platform profiling that inherently bypasses current platform safeguards, highlighting a systemic vulnerability in the ad ecosystem and the urgent need for responsible web AI governance in the generative AI era. The code is available at https://github.com/Breezelled/when-ads-become-profiles.

cs.HC↗

arXiv2025

Exploring the Alignment of Perceived and Measured Sleep Quality with Working Memory using Consumer Wearables

Wearable devices offer detailed sleep-tracking data. However, whether this information enhances our understanding of sleep or simply quantifies already-known patterns remains unclear. This work explores the relationship between subjective sleep self-assessments and sensor data from an Oura ring over 4--8 weeks in-the-wild. 29 participants rated their sleep quality daily compared to the previous night and completed a working memory task. Our findings reveal that differences in REM sleep, nocturnal heart rate, N-Back scores, and bedtimes highly predict sleep self-assessment in significance and effect size. For N-Back performance, REM sleep duration, prior night's REM sleep, and sleep self-assessment are the strongest predictors. We demonstrate that self-report sensitivity towards sleep markers differs among participants. We identify three groups, highlighting that sleep trackers provide more information gain for some users than others. Additionally, we make all experiment data publicly available.

cs.HC↗

arXiv2025

AlphaPIG: The Nicest Way to Prolong Interactive Gestures in Extended Reality

Mid-air gestures serve as a common interaction modality across Extended Reality (XR) applications, enhancing engagement and ownership through intuitive body movements. However, prolonged arm movements induce shoulder fatigue, known as "Gorilla Arm Syndrome", degrading user experience and reducing interaction duration. Although existing ergonomic techniques derived from Fitts' law (such as reducing target distance, increasing target width, and modifying control-display gain) provide some fatigue mitigation, their implementation in XR applications remains challenging due to the complex balance between user engagement and physical exertion. We present AlphaPIG, a meta-technique designed to Prolong Interactive Gestures by leveraging real-time fatigue predictions. AlphaPIG assists designers in extending and improving XR interactions by enabling automated fatigue-based interventions. Through adjustment of intervention timing and intensity decay rate, designers can explore and control the trade-off between fatigue reduction and potential effects such as decreased body ownership. We validated AlphaPIG's effectiveness through a study (N=22) implementing the widely-used Go-Go technique. Results demonstrated that AlphaPIG significantly reduces shoulder fatigue compared to non-adaptive Go-Go, while maintaining comparable perceived body ownership and agency. Based on these findings, we discuss positive and negative perceptions of the intervention. By integrating real-time fatigue prediction with adaptive intervention mechanisms, AlphaPIG constitutes a critical first step towards creating fatigue-aware applications in XR.

cs.HC↗

arXiv2025

A Review of Cognitive Readiness, Wearable Devices, and Prospects

In Human-Computer Interaction (HCI) and Ubiquitous Computing, the objective of optimizing device interactions and personalizing user experiences has placed a new emphasis on accurately evaluating cognitive readiness using wearable devices. Interpreting cognitive readiness in real-world scenarios is complex due to the plethora of potential physiological measures, individual variability, and the limitations of wearable devices. In this review, we present a systematic overview of key physiological measures that can be used for an in-depth assessment of cognitive readiness. These measures can serve as proxies for detailed assessments of cognitive readiness. This review serves as a tool for assessing cognitive readiness for diverse applications, with special focus on in-the-wild research settings. In addition, due to the complexity of measurements and devices, we propose the development of robust catalog for cognitive readiness measurements.

cs.HC↗

arXiv2024

AR-Facilitated Safety Inspection and Fall Hazard Detection on Construction Sites

Together with industry experts, we are exploring the potential of head-mounted augmented reality to facilitate safety inspections on high-rise construction sites. A particular concern in the industry is inspecting perimeter safety screens on higher levels of construction sites, intended to prevent falls of people and objects. We aim to support workers performing this inspection task by tracking which parts of the safety screens have been inspected. We use machine learning to automatically detect gaps in the perimeter screens that require closer inspection and remediation and to automate reporting. This work-in-progress paper describes the problem, our early progress, concerns around worker privacy, and the possibilities to mitigate these.

cs.HC↗

arXiv2024

Inside Out or Not: Privacy Implications of Emotional Disclosure

Privacy is dynamic, sensitive, and contextual, much like our emotions. Previous studies have explored the interplay between privacy and context, privacy and emotion, and emotion and context. However, there remains a significant gap in understanding the interplay of these aspects simultaneously. In this paper, we present a preliminary study investigating the role of emotions in driving individuals' information sharing behaviour, particularly in relation to urban locations and social ties. We adopt a novel methodology that integrates context (location and time), emotion, and personal information sharing behaviour, providing a comprehensive analysis of how contextual emotions affect privacy. The emotions are assessed with both self-reporting and electrodermal activity (EDA). Our findings reveal that self-reported emotions influence personal information-sharing behaviour with distant social groups, while neutral emotions lead individuals to share less precise information with close social circles, a pattern is potentially detectable with wrist-worn EDA. Our study helps lay the foundation for personalised emotion-aware strategies to mitigate oversharing risks and enhance user privacy in the digital age.

cs.CY↗

arXiv2024

NICER: A New and Improved Consumed Endurance and Recovery Metric to Quantify Muscle Fatigue of Mid-Air Interactions

Natural gestures are crucial for mid-air interaction, but predicting and managing muscle fatigue is challenging. Existing torque-based models are limited in their ability to model above-shoulder interactions and to account for fatigue recovery. We introduce a new hybrid model, NICER, which combines a torque-based approach with a new term derived from the empirical measurement of muscle contraction and a recovery factor to account for decreasing fatigue during rest. We evaluated NICER in a mid-air selection task using two interaction methods with different degrees of perceived fatigue. Results show that NICER can accurately model above-shoulder interactions as well as reflect fatigue recovery during rest periods. Moreover, both interaction methods show a stronger correlation with subjective fatigue measurement (r = 0.978/0.976) than a previous model, Cumulative Fatigue (r = 0.966/ 0.923), confirming that NICER is a powerful analytical tool to predict fatigue across a variety of gesture-based interactive applications.

cs.HC↗

arXiv2024

Identifying Periods of Cyclical Stress in University Students Using Wearables In-the-Wild

University students encounter various forms of stress during their academic journey, including cyclical stress associated with final exams. Supporting their well-being means helping them manage their stress levels. In this study, we used a wearable health-tracking ring on a cohort of 103 Japanese university students for up to 28 months in the wild. The study aimed to investigate whether group-wide biomarkers of stress can be identified in a sample having similar daily schedules and whether these occurrences can be pinpointed to specific periods of the academic year. We found population-wide increased stress markers during exams, New Year's, and job hunting season, a Japanese job market peculiarity. Our results highlight the available potential of unobtrusive, in-situ detection of the current mental state of university student populations using off-the-shelf wearables from noisy data, with significant implications for the well-being of the users. Our approach and method of analysis allows for monitoring the student body's stress level without singling out individuals and therefore represents a privacy-preserving method. This way, new and sudden stress increases can be recognized, which can help identify the stressor and inform the design and introduction of counter measures.

cs.CY↗

arXiv2023

PhysioCHI: Towards Best Practices for Integrating Physiological Signals in HCI

Recently, we saw a trend toward using physiological signals in interactive systems. These signals, offering deep insights into users' internal states and health, herald a new era for HCI. However, as this is an interdisciplinary approach, many challenges arise for HCI researchers, such as merging diverse disciplines, from understanding physiological functions to design expertise. Also, isolated research endeavors limit the scope and reach of findings. This workshop aims to bridge these gaps, fostering cross-disciplinary discussions on usability, open science, and ethics tied to physiological data in HCI. In this workshop, we will discuss best practices for embedding physiological signals in interactive systems. Through collective efforts, we seek to craft a guiding document for best practices in physiological HCI research, ensuring that it remains grounded in shared principles and methodologies as the field advances.

cs.HC↗

arXiv2023

Context-Dependent Memory in Situated Visualization

Situated visualization presents data alongside their source context (physical referent). While environmental factors influence memory recall (known as Context-Dependent Memory or CDM), how physical context affects cognition in real-world tasks such as working with visualizations in situated contexts is unclear. This study explores the design space of information memorability in situated visualization through the lens of CDM. We investigate the presence of physical referents for creating contextual cues in desktop and Virtual Reality (VR) environments. Across three studies (n=144), we observe a trend suggesting a CDM effect due to contextual referent is more apparent in VR. Overall, we did not find statistically significant evidence of a CDM effect due to the presence of a referent. However, we did find a significant CDM effect for lighting conditions. This suggests that representing the entire environment, rather than the physical objects alone, may be necessary to provide sufficiently strong contextual memory cues.

cs.HC↗