arXiv subjects

Cathy Mengying Fang

Publications and source records attributed to Cathy Mengying Fang.

17 recordsLinked to original sources

arXiv2026

AI emotional support is better only when chosen, but shifts preferences even when it is not

People increasingly face a novel decision when seeking emotional support: human or AI. In existing studies, AI's empathic messages are rated as well as or better than humans'. But these studies either assigned the support source or honored people's choice. In real life, support is often incongruent with choice, as people want one source and receive the other. Across three experiments (N = 1,951), participants chose whether to share an emotional experience with a human or an AI, then were randomly assigned to a congruent or incongruent partner. AI support was rated as superior only among those who had chosen it. Yet regardless of congruence, interacting with AI increased willingness to choose it again. In a 28-day study with OpenAI (N = 981), daily conversations shifted preferences toward AI and away from humans, but only when conversations turned personal. Emotional support choices are thus path-dependent, progressively redirecting away from human connection.

cs.AI↗

arXiv2026

Stumbling Into AI Emotional Dependence: How Routine AI Interactions Reshape Human Connection

Public discourse and emerging policy typically assume that AI emotional support is a deliberate act: a lonely user consciously seeking comfort from a dedicated companion chatbot. In this paper, we draw on emerging empirical evidence and argue that this picture is inaccurate on two accounts, both in how AI emotional support arises and how it shapes future behavior. First, AI emotional support commonly emerges incidentally within task-oriented interactions on general-purpose platforms, much as workplace friendships deepen through collaboration. Second, these incidental encounters are path-dependent: positive experiences of AI emotional support update people's beliefs about AI's emotional capabilities and redirect their choices for future emotional support, increasing preference for AI and decreasing preference for humans. We review recent evidence, including a large-scale longitudinal study conducted in collaboration with OpenAI, showing that daily five-minute conversations with an AI about personal issues over 28 days led to a 10.3% decrease in the preference for seeking support from humans and an 11.6% increase in the preference for AI. These findings suggest that current policy, focused on companion apps and isolated interactions, cannot adequately protect human connection. Instead, effective regulations should extend to general-purpose AI systems and address cumulative, trajectory-level changes in how people seek support. Recognizing how people stumble into AI emotional support and how those encounters redirect human connections over time is essential to safeguarding human well-being.

cs.AI↗

arXiv2026

ThoughtTrace: Understanding User Thoughts in Real-World LLM Interactions

Conversational AI has now reached billions of users, yet existing datasets capture only what people say, not what they think. We introduce ThoughtTrace, the first large-scale dataset that pairs real-world multi-turn human--AI conversations with users' self-reported thoughts: their reasons for sending prompts and reactions to assistant responses. ThoughtTrace comprises 1,058 users, 2,155 conversations, 17,058 turns, and 10,174 thought annotations collected across 20 language models. Our analysis shows that ThoughtTrace captures long-horizon, topically diverse interactions, and that thoughts are semantically distinct from messages, difficult for frontier LLMs to infer from context, diverse in content, and tied to conversation stages. We further demonstrate the utility of thoughts for downstream modeling. First, thoughts improve user-behavior prediction as inference-time context. Second, thought-guided rewrites provide fine-grained alignment signals for training personalized assistants. Together, ThoughtTrace establishes user thoughts as a new data modality for studying the cognitive dynamics behind human--AI interaction and provides a foundation for building assistants that better understand and adapt to users' latent goals, preferences, and needs.

cs.CL↗

arXiv2026

AI-Wrapped: Participatory, Privacy-Preserving Measurement of Longitudinal LLM Use In-the-Wild

Alignment research on large language models (LLMs) increasingly depends on understanding how these systems are used in everyday contexts. Yet naturalistic interaction data is difficult to access due to privacy constraints and platform control. We present AI-Wrapped, a prototype workflow for collecting naturalistic LLM chatbot usage data while providing participants with an immediate "wrapped"-style report on their usage statistics, top topics, and behavioral patterns. We report findings from an initial deployment with 82 U.S.-based adults across 48,495 conversations from their 2025 chat histories. Participants used LLMs for both instrumental and reflective purposes and had topics with emotional or existential themes. Some usage patterns reflect potential over-reliance or perfectionism. Heavy users showed comparatively more reflective exchanges than primarily transactional ones. Methodologically, even with zero data retention and PII removal, participants may remain hesitant to share chat data due to perceived privacy and judgment risks, underscoring the importance of transparent design when building measurement infrastructure for alignment research.

cs.HC↗

arXiv2025

How AI and Human Behaviors Shape Psychosocial Effects of Extended Chatbot Use: A Longitudinal Randomized Controlled Study

As people increasingly seek emotional support and companionship from AI chatbots, understanding how such interactions impact mental well-being becomes critical. We conducted a four-week randomized controlled experiment (n=981, >300k messages) to investigate how interaction modes (text, neutral voice, and engaging voice) and conversation types (open-ended, non-personal, and personal) influence four psychosocial outcomes: loneliness, social interaction with real people, emotional dependence on AI, and problematic AI usage. No significant effects were detected from experimental conditions, despite conversation analyses revealing differences in AI and human behavioral patterns across the conditions. Instead, participants who voluntarily used the chatbot more, regardless of assigned condition, showed consistently worse outcomes. Individuals' characteristics, such as higher trust and social attraction towards the AI chatbot, are associated with higher emotional dependence and problematic use. These findings raise deeper questions about how artificial companions may reshape the ways people seek, sustain, and substitute human connections.

cs.HC↗

arXiv2025

EmoSign: A Multimodal Dataset for Understanding Emotions in American Sign Language

Unlike spoken languages where the use of prosodic features to convey emotion is well studied, indicators of emotion in sign language remain poorly understood, creating communication barriers in critical settings. Sign languages present unique challenges as facial expressions and hand movements simultaneously serve both grammatical and emotional functions. To address this gap, we introduce EmoSign, the first sign video dataset containing sentiment and emotion labels for 200 American Sign Language (ASL) videos. We also collect open-ended descriptions of emotion cues. Annotations were done by 3 Deaf ASL signers with professional interpretation experience. Alongside the annotations, we include baseline models for sentiment and emotion classification. This dataset not only addresses a critical gap in existing sign language research but also establishes a new benchmark for understanding model capabilities in multimodal emotion recognition for sign languages. The dataset is made available at https://huggingface.co/datasets/catfang/emosign.

cs.CV↗

arXiv2025

Perspectives on Capturing Emotional Expressiveness in Sign Language

Significant advances have been made in our ability to understand and generate emotionally expressive content such as text and speech, yet comparable progress in sign language technologies remain limited. While computational approaches to sign language translation have focused on capturing lexical content, the emotional dimensions of sign language communication remain largely unexplored. Through semi-structured interviews with eight sign language users across Singapore, Sri Lanka and the United States, including both Deaf and Hard of hearing (DHH) and hearing signers, we investigate how emotions are expressed and perceived in sign languages. Our findings highlight the role of both manual and non-manual elements in emotional expression, revealing universal patterns as well as individual and cultural variations in how signers communicate emotions. We identify key challenges in capturing emotional nuance for sign language translation, and propose design considerations for developing more emotionally-aware sign language technologies. This work contributes to both theoretical understanding of emotional expression in sign language and practical development of interfaces to better serve diverse signing communities.

cs.HC↗

arXiv2025

The Goldilocks Time Window for Proactive Interventions in Wearable AI Systems

As AI systems become increasingly integrated into our daily lives and into wearable form factors, there's a fundamental tension between their potential to proactively assist us and the risk of creating intrusive, dependency-forming experiences. This work proposes the concept of a Goldilocks Time Window -- a contextually adaptive time window for proactive AI systems to deliver effective interventions. We discuss the critical factors that determine the time window, and the need of a framework for designing and evaluating proactive AI systems that can navigate this tension successfully.

cs.HC↗

arXiv2025

Leveraging AI-Generated Emotional Self-Voice to Nudge People towards their Ideal Selves

Emotions, shaped by past experiences, significantly influence decision-making and goal pursuit. Traditional cognitive-behavioral techniques for personal development rely on mental imagery to envision ideal selves, but may be less effective for individuals who struggle with visualization. This paper introduces Emotional Self-Voice (ESV), a novel system combining emotionally expressive language models and voice cloning technologies to render customized responses in the user's own voice. We investigate the potential of ESV to nudge individuals towards their ideal selves in a study with 60 participants. Across all three conditions (ESV, text-only, and mental imagination), we observed an increase in resilience, confidence, motivation, and goal commitment, and the ESV condition was perceived as uniquely engaging and personalized. We discuss the implications of designing generated self-voice systems as a personalized behavioral intervention for different scenarios.

cs.HC↗

arXiv2025

Investigating Affective Use and Emotional Well-being on ChatGPT

As AI chatbots see increased adoption and integration into everyday life, questions have been raised about the potential impact of human-like or anthropomorphic AI on users. In this work, we investigate the extent to which interactions with ChatGPT (with a focus on Advanced Voice Mode) may impact users' emotional well-being, behaviors and experiences through two parallel studies. To study the affective use of AI chatbots, we perform large-scale automated analysis of ChatGPT platform usage in a privacy-preserving manner, analyzing over 3 million conversations for affective cues and surveying over 4,000 users on their perceptions of ChatGPT. To investigate whether there is a relationship between model usage and emotional well-being, we conduct an Institutional Review Board (IRB)-approved randomized controlled trial (RCT) on close to 1,000 participants over 28 days, examining changes in their emotional well-being as they interact with ChatGPT under different experimental settings. In both on-platform data analysis and the RCT, we observe that very high usage correlates with increased self-reported indicators of dependence. From our RCT, we find that the impact of voice-based interactions on emotional well-being to be highly nuanced, and influenced by factors such as the user's initial emotional state and total usage duration. Overall, our analysis reveals that a small number of users are responsible for a disproportionate share of the most affective cues.

cs.HC↗

arXiv2025

Mirai: A Wearable Proactive AI "Inner-Voice" for Contextual Nudging

People often find it difficult to turn their intentions into real actions -- a challenge that affects both personal growth and mental well-being. While established methods like cognitive-behavioral therapy and mindfulness training help people become more aware of their behaviors and set clear goals, these approaches cannot provide immediate guidance when people fall into automatic reactions or habits. We introduce Mirai, a novel wearable AI system with an integrated camera, real-time speech processing, and personalized voice-cloning to provide proactive and contextual nudges for positive behavior change. Mirai continuously monitors and analyzes the user's environment to anticipate their intentions, generating contextually-appropriate responses delivered in the user's own cloned voice. We demonstrate the application of Mirai through three scenarios focusing on dietary choices, work productivity, and communication skills. We also discuss future work on improving the proactive agent via human feedback and the need for a longitudinal study in naturalistic settings.

cs.HC↗

arXiv2024

Enabling Waypoint Generation for Collaborative Robots using LLMs and Mixed Reality

Programming a robotic is a complex task, as it demands the user to have a good command of specific programming languages and awareness of the robot's physical constraints. We propose a framework that simplifies robot deployment by allowing direct communication using natural language. It uses large language models (LLM) for prompt processing, workspace understanding, and waypoint generation. It also employs Augmented Reality (AR) to provide visual feedback of the planned outcome. We showcase the effectiveness of our framework with a simple pick-and-place task, which we implement on a real robot. Moreover, we present an early concept of expressive robot behavior and skill generation that can be used to communicate with the user and learn new skills (e.g., object grasping).

cs.HC↗

arXiv2024

Purrfect Pitch: Exploring Musical Interval Learning through Multisensory Interfaces

We introduce Purrfect Pitch, a system consisting of a wearable haptic device and a custom-designed learning interface for musical ear training. We focus on the ability to identify musical intervals (sequences of two musical notes), which is a perceptually ambiguous task that usually requires strenuous rote training. With our system, the user would hear a sequence of two tones while simultaneously receiving two corresponding vibrotactile stimuli on the back. Providing haptic feedback along the back makes the auditory distance between the two tones more salient, and the back-worn design is comfortable and unobtrusive. During training, the user receives multi-sensory feedback from our system and inputs their guessed interval value on our web-based learning interface. They see a green (otherwise red) screen for a correct guess with the correct interval value. Our study with 18 participants shows that our system enables novice learners to identify intervals more accurately and consistently than those who only received audio feedback, even after the haptic feedback is removed. We also share further insights on how to design a multisensory learning system.

cs.HC↗

arXiv2024

PhysioLLM: Supporting Personalized Health Insights with Wearables and Large Language Models

We present PhysioLLM, an interactive system that leverages large language models (LLMs) to provide personalized health understanding and exploration by integrating physiological data from wearables with contextual information. Unlike commercial health apps for wearables, our system offers a comprehensive statistical analysis component that discovers correlations and trends in user data, allowing users to ask questions in natural language and receive generated personalized insights, and guides them to develop actionable goals. As a case study, we focus on improving sleep quality, given its measurability through physiological data and its importance to general well-being. Through a user study with 24 Fitbit watch users, we demonstrate that PhysioLLM outperforms both the Fitbit App alone and a generic LLM chatbot in facilitating a deeper, personalized understanding of health data and supporting actionable steps toward personal health goals.

cs.HC↗

arXiv2024

LLMR: Real-time Prompting of Interactive Worlds using Large Language Models

We present Large Language Model for Mixed Reality (LLMR), a framework for the real-time creation and modification of interactive Mixed Reality experiences using LLMs. LLMR leverages novel strategies to tackle difficult cases where ideal training data is scarce, or where the design goal requires the synthesis of internal dynamics, intuitive analysis, or advanced interactivity. Our framework relies on text interaction and the Unity game engine. By incorporating techniques for scene understanding, task planning, self-debugging, and memory management, LLMR outperforms the standard GPT-4 by 4x in average error rate. We demonstrate LLMR's cross-platform interoperability with several example worlds, and evaluate it on a variety of creation and modification tasks to show that it can produce and edit diverse objects, tools, and scenes. Finally, we conducted a usability study (N=11) with a diverse set that revealed participants had positive experiences with the system and would use it again.

cs.HC↗

arXiv2024

Real-time Animation Generation and Control on Rigged Models via Large Language Models

We introduce a novel method for real-time animation control and generation on rigged models using natural language input. First, we embed a large language model (LLM) in Unity to output structured texts that can be parsed into diverse and realistic animations. Second, we illustrate LLM's potential to enable flexible state transition between existing animations. We showcase the robustness of our approach through qualitative results on various rigged models and motions.

cs.GR↗

arXiv2023

VR Haptics at Home: Repurposing Everyday Objects and Environment for Casual and On-Demand VR Haptic Experiences

This paper introduces VR Haptics at Home, a method of repurposing everyday objects in the home to provide casual and on-demand haptic experiences. Current VR haptic devices are often expensive, complex, and unreliable, which limits the opportunities for rich haptic experiences outside research labs. In contrast, we envision that, by repurposing everyday objects as passive haptics props, we can create engaging VR experiences for casual uses with minimal cost and setup. To explore and evaluate this idea, we conducted an in-the-wild study with eight participants, in which they used our proof-of-concept system to turn their surrounding objects such as chairs, tables, and pillows at their own homes into haptic props. The study results show that our method can be adapted to different homes and environments, enabling more engaging VR experiences without the need for complex setup process. Based on our findings, we propose a possible design space to showcase the potential for future investigation.

cs.HC↗