How do people plan digitally: An in-the-wild investigation of task planning through a smartphone app
研究通过分析24,265项任务及957名用户数据，探讨了人们如何使用智能手机应用进行日常任务规划，并揭示了不同用户的参与度模式。
0 citationsRead paper
Selected work
Representative Papers
Social Determinants of Health and Fentanyl Overdose Mortality Across US Counties: An XGBoost and SHAP Analysis Identifying Silent Risk Counties and Treatment Deserts
This study investigates the social and structural determinants driving rising fentanyl overdose mortality across U.S. counties, with a focus on identifying “silent-risk” counties—those not yet exhibiting high death rates but possessing significant underlying vulnerability—and “treatment deserts,” characterized by severe shortages of addiction treatment resources. Integrating four major governmental datasets, including CDC records, the authors apply interpretable machine learning (XGBoost with SHAP values) to 2022 county-level data, complemented by five-fold cross-validation, spatial autocorrelation analysis (Moran’s I), K-means clustering, and standardized mortality ratios (SMRs). The model demonstrates strong performance (Spearman ρ = 0.67, R² = 0.457), successfully flagging 143 silent-risk counties. Treatment desert counties exhibit 52.6% higher overdose mortality, while spatial analysis reveals 75 hotspots and 136 coldspots; notably, 72% of suppressed-data counties are rural and 65% are treatment deserts, underscoring the tight linkage between social determinants and spatial clustering patterns.
0 citationsRead paper
StoryComposerAI: Supporting Human-AI Story Co-Creation Through Decomposition and Linking
This work addresses the challenge users face in maintaining narrative coherence and visual consistency when editing localized elements in human-AI collaborative digital storytelling. To this end, the authors propose a novel “Creative Decomposition and Linking” paradigm that explicitly decomposes a story into structured units—such as plot, characters, and scenes—and leverages generative AI to enable independent yet controllable generation of each unit through structured prompts and cross-modal association modeling, all while enforcing global coherence constraints. The resulting StoryComposerAI system significantly enhances users’ fine-grained control over the creative process while preserving narrative continuity and multimodal consistency, thereby demonstrating the effectiveness and innovation of the proposed paradigm.
0 citationsRead paper
Creating Disability Story Videos with Generative AI: Motivation, Expression, and Sharing
This study addresses the challenges generative AI poses in faithfully representing the lived experiences and identities of disabled individuals when supporting their creation of personal narratives, often due to bias and expressive distortion. Grounded in digital storytelling theory, the research collaborates with nine disabled participants to propose a “Critical Moment Rendering” analytical framework. This framework systematically examines participants’ motivations, practices, and experiences in using generative AI for video creation through four dimensions: the re-presentation of elusive scenes, the negotiation of identity concealment and disclosure, contextual coherence, and emotional expression. The findings yield design implications for AI systems tailored to disabled storytelling, emphasizing narrative integrity, adaptive media formatting, and robust error-correction mechanisms, thereby offering both theoretical and practical foundations for developing more inclusive generative AI tools.
0 citationsRead paper
LLM-Based Multi-Agent System for Simulating and Analyzing Marketing and Consumer Behavior
Traditional consumer behavior modeling relies on post-hoc analysis and rule-based agent models, limiting its capacity to capture cognitive complexity and emergent social dynamics. To address this, we propose a large language model (LLM)-driven generative multi-agent system that abandons predefined rules and instead enables dynamic simulation of consumer decision-making, habit formation, and social diffusion through natural-language interaction, internal cognitive modeling, and co-evolutionary learning. Deployed in a price-promotion sandbox environment, the system autonomously generates interpretable strategic feedback and— for the first time—uncovers latent population-level consumption patterns and social cascade effects. Compared to conventional approaches, our framework achieves higher ecological validity, lower experimental cost, and superior scalability, establishing a novel computational experimentation paradigm for pre-testing marketing strategies.
0 citationsRead paper
Recent publications
Latest Papers
How do people plan digitally: An in-the-wild investigation of task planning through a smartphone app
研究通过分析24,265项任务及957名用户数据，探讨了人们如何使用智能手机应用进行日常任务规划，并揭示了不同用户的参与度模式。
0 citationsRead paper
Social Determinants of Health and Fentanyl Overdose Mortality Across US Counties: An XGBoost and SHAP Analysis Identifying Silent Risk Counties and Treatment Deserts
This study investigates the social and structural determinants driving rising fentanyl overdose mortality across U.S. counties, with a focus on identifying “silent-risk” counties—those not yet exhibiting high death rates but possessing significant underlying vulnerability—and “treatment deserts,” characterized by severe shortages of addiction treatment resources. Integrating four major governmental datasets, including CDC records, the authors apply interpretable machine learning (XGBoost with SHAP values) to 2022 county-level data, complemented by five-fold cross-validation, spatial autocorrelation analysis (Moran’s I), K-means clustering, and standardized mortality ratios (SMRs). The model demonstrates strong performance (Spearman ρ = 0.67, R² = 0.457), successfully flagging 143 silent-risk counties. Treatment desert counties exhibit 52.6% higher overdose mortality, while spatial analysis reveals 75 hotspots and 136 coldspots; notably, 72% of suppressed-data counties are rural and 65% are treatment deserts, underscoring the tight linkage between social determinants and spatial clustering patterns.
0 citationsRead paper
StoryComposerAI: Supporting Human-AI Story Co-Creation Through Decomposition and Linking
This work addresses the challenge users face in maintaining narrative coherence and visual consistency when editing localized elements in human-AI collaborative digital storytelling. To this end, the authors propose a novel “Creative Decomposition and Linking” paradigm that explicitly decomposes a story into structured units—such as plot, characters, and scenes—and leverages generative AI to enable independent yet controllable generation of each unit through structured prompts and cross-modal association modeling, all while enforcing global coherence constraints. The resulting StoryComposerAI system significantly enhances users’ fine-grained control over the creative process while preserving narrative continuity and multimodal consistency, thereby demonstrating the effectiveness and innovation of the proposed paradigm.
0 citationsRead paper
Creating Disability Story Videos with Generative AI: Motivation, Expression, and Sharing
This study addresses the challenges generative AI poses in faithfully representing the lived experiences and identities of disabled individuals when supporting their creation of personal narratives, often due to bias and expressive distortion. Grounded in digital storytelling theory, the research collaborates with nine disabled participants to propose a “Critical Moment Rendering” analytical framework. This framework systematically examines participants’ motivations, practices, and experiences in using generative AI for video creation through four dimensions: the re-presentation of elusive scenes, the negotiation of identity concealment and disclosure, contextual coherence, and emotional expression. The findings yield design implications for AI systems tailored to disabled storytelling, emphasizing narrative integrity, adaptive media formatting, and robust error-correction mechanisms, thereby offering both theoretical and practical foundations for developing more inclusive generative AI tools.
0 citationsRead paper
LLM-Based Multi-Agent System for Simulating and Analyzing Marketing and Consumer Behavior
Traditional consumer behavior modeling relies on post-hoc analysis and rule-based agent models, limiting its capacity to capture cognitive complexity and emergent social dynamics. To address this, we propose a large language model (LLM)-driven generative multi-agent system that abandons predefined rules and instead enables dynamic simulation of consumer decision-making, habit formation, and social diffusion through natural-language interaction, internal cognitive modeling, and co-evolutionary learning. Deployed in a price-promotion sandbox environment, the system autonomously generates interpretable strategic feedback and— for the first time—uncovers latent population-level consumption patterns and social cascade effects. Compared to conventional approaches, our framework achieves higher ecological validity, lower experimental cost, and superior scalability, establishing a novel computational experimentation paradigm for pre-testing marketing strategies.
0 citationsRead paper