Conversational assistants, such as Siri, Alexa, and Google Assistant, are ubiquitous beginning to be used portals for medical services. However, the potential safety issues of using conversational assistants information by patients consumers not understood.To determine prevalence nature harm that could result from or information.Participants were given problems pose asked an action take based on system. Assignment tasks systems randomized across participants, participants queried in their...

The use of retrieval-augmented generation (RAG) to retrieve relevant information from an external knowledge source enables large language models (LLMs) answer questions over private and/or previously unseen document collections. However, RAG fails on global directed at entire text corpus, such as "What are the main themes in dataset?", since this is inherently a query-focused summarization (QFS) task, rather than explicit retrieval task. Prior QFS methods, meanwhile, fail scale quantities...

Rehearsing in front of a live audience is invaluable when preparing for important presentations. However, not all presenters take the opportunity to engage such rehearsal, due time constraints, availability listeners who can provide constructive feedback, or public speaking anxiety. We present RoboCOP, an automated anthropomorphic robot head that acts as coach spoken feedback during presentation rehearsals at both individual slide and overall level. The offers conversational coaching on...

When deployed on mobile devices, virtual agents have the potential to deliver advice regarding medical conditions, as well provide a ubiquitous channel for health education and behavior change variety of chronic conditions. We describe design guidelines agent dialogues support disease management, general-purpose smartphone-based architecture conversational that simulates face-to-face counseling conversations with patients, an initial implementation provides patients heart condition atrial...

In this paper, we introduce the concept of principal communities and propose a graph encoder embedding method that concurrently detects these achieves vertex embedding. Given adjacency matrix with labels, computes sample community score for each community, ranking them to measure importance estimate set communities. The then produces by retaining only dimensions corresponding Theoretically, define population version based on random Bernoulli distribution. We prove preserves conditional...

Analyzing large-scale time-series network data, such as social media and email communications, poses a significant challenge in understanding dynamics, detecting anomalies, predicting trends. In particular, the scalability of graph analysis is critical hurdle impeding progress downstream inference. To address this challenge, we introduce temporal encoder embedding method. This approach leverages ground-truth or estimated vertex labels, enabling an efficient data processing billions edges...

The quality of most professional oral presentations is often poor, owing to a number factors, including public speaking anxiety. We present DynamicDuo, system that uses an automated, life-sized, animated agent help inexperienced speakers deliver their in front audience. design the was informed by analysis TED talks given two human presenters identify common dual-presentation formats and transition behaviors used. In within-subjects study (N=12) comparing co-presenting with DynamicDuo against...

Public speaking is an essential task for success in many careers, yet fear of public makes this undesirable activity most people and negatively affects the quality presentations. Cognitive behavioral therapy effective tool helping overcome social anxieties disorders, including anxiety. However, do not engage anxiety, due to time constraints other barriers. We present a virtual coach that uses cognitive techniques help presenters restructure irrational thoughts associated with The design was...

Most persons with spinal cord injury (SCI) require training and support for self-care management to help prevent the development of serious secondary conditions after hospital discharge. We have designed a virtual coach system, in which an animated character engages users simulated face-to-face conversation provide health education motivate healthy behavior. conducted exploratory study nine participants who SCI examine acceptance attitudes towards our system. Results show that are highly...

Rehearsal is a critical component of preparing to give an oral presentation, yet it frequently abbreviated, performed in ways that are inefficient or ineffective, simply omitted. We conducted exploratory study understand the relationship between theory and practice presentation rehearsal, classifying our qualitative results into five themes motivate more structured rehearsal support deeply integrated slide software. In within-subject (N=12) comparing against participants' existing practices,...

Longitudinal agent-based interventions only work if people continue using them on a regular basis, thus identifying users who are at risk of disengaging from these applications is important for retention and efficacy. We develop machine learning models that predict long-term user engagement in three longitudinal virtual health interventions. achieve accuracies 74% to 90% predicting dropout given prediction period the intervention based user's past interactions with agent. Our contain...

We present Friendly Face - a virtual agent designed to reduce public speaking anxiety by standing within an audience. The senses the speaker's behavior during oral presentation and provides emotional instrumental support. system unobtrusively tracks motion, speech, prosody of presenter intuitive interface give supportive feedback whenever looks at agent, attentive listening through gaze backchannel behavior, time topic cueing based on real-time analysis speech content compared slide...

Prior studies of public speaking behavior have demonstrated that anxiety monotonically decreases with the number co-presenters giving an oral presentation and increases size audience. However, speaker "hand off" behavior---the verbal nonverbal cues used to transition from one another---and its effect on quality has not been systematically studied. In this work we report two empirical hand-off during human co-presentations. We find for hand-offs prepared rehearsed presentations differ...

This paper proposes using face as a touch surface and employing hand-over-face (HOF) gestures novel input modality for interaction with smartphones, especially when is limited. We contribute InterFace, general system framework that enables the HOF advanced computer vision techniques. As an examplar of usage this framework, we demonstrate feasibility usefulness Android application improving single-user group selfie-taking experience through providing appearance customization in real-time. In...

The fear of forgetting what to say next is a common problem in oral presentation delivery. To maintain good content coverage and reduce anxiety, many presenters use written notes conjunction with slides. However, excessive during delivery can lead disengagement the audience low quality presentations. We designed IntelliPrompter, speech-based note display system that automatically tracks presenter's each slide's dynamically adjusts interface highlight most likely topic present. developed two...

The high incidence of literacy deficits among people with severe speech impairments (SSI) has been well documented. Without skills, SSI are unable to effectively use orthographic-based communication systems generate novel linguistic items in spontaneous conversation. To address this problem, phoneme-based have proposed which enable users create spoken output from phoneme sequences. In paper, we investigate whether prediction techniques can be employed improve the usability such systems. We...