Embedded Files
VIDEO-IntergenerationalUnderstandingRoom (1).mp4AI Simulation for Historical Interpretation and Education (2025-)
Abstract: When historian Steven Johnson struggled to plot a turning point in his latest book, he turned to NotebookLM, an AI tool that let him interrogate sources he had personally curated. As Bill Wasik (2025) discusses in the New York Times Magazine, what emerged was not a replacement for historical scholarship but a new kind of collaboration—one in which AI expanded the horizons of interpretation. This captures a potential shift: history is beginning to be told with AI as an active partner in the narrative process. How might the practice of history change with the involvement of artificial intelligence? What would our histories look like if AI became an active “storyteller”? In this project, we address these questions through a multi-agent AI simulation approach. We use China's Cultural Revolution as one case study. We examine: 1) What narratives of the Cultural Revolution emerge from AI-driven simulations? and 2) How do narrative features of AI simulations (e.g., agent tone, linguistic style, and dialogue structure) shape audience engagement? Further, how does exposure to multiple, sometimes conflicting, AI personas influence audiences’ evaluation of historical narratives?
Abstract: When historian Steven Johnson struggled to plot a turning point in his latest book, he turned to NotebookLM, an AI tool that let him interrogate sources he had personally curated. As Bill Wasik (2025) discusses in the New York Times Magazine, what emerged was not a replacement for historical scholarship but a new kind of collaboration—one in which AI expanded the horizons of interpretation. This captures a potential shift: history is beginning to be told with AI as an active partner in the narrative process. How might the practice of history change with the involvement of artificial intelligence? What would our histories look like if AI became an active “storyteller”? In this project, we address these questions through a multi-agent AI simulation approach. We use China's Cultural Revolution as one case study. We examine: 1) What narratives of the Cultural Revolution emerge from AI-driven simulations? and 2) How do narrative features of AI simulations (e.g., agent tone, linguistic style, and dialogue structure) shape audience engagement? Further, how does exposure to multiple, sometimes conflicting, AI personas influence audiences’ evaluation of historical narratives?
AI Adoption and Integration in Digital Humanities (2024-)
Preprint: https://arxiv.org/abs/2404.12458
Abstract: The advent of generative artificial intelligence (GenAI) technologies is changing the research landscape and potentially has significant implications for Digital Humanities (DH)–a field inherently intertwined with technologies. This project investigates how DH scholars adopt and critically evaluate GenAI technologies for their research. We demonstrate that DH research communities hold divided opinions and differing imaginaries about the role of GenAI in DH scholarship. While scholars acknowledge the benefits of GenAI in enhancing research efficiency and enabling reskilling, many remain concerned about its potential to disrupt their intellectual identities. Situated within the history of DH and viewed through the lens of Actor-Network Theory, our findings suggest that the enrolment of GenAI technology, alongside other human and non-human actors within an integrated network, is gradually changing the field, though this transformation remains contested, shaped by ongoing negotiations. Our study is one of the first empirical analyses of this topic and has the potential to serve as a building block for future inquiries into the impact of GenAI on DH scholarship.
CCLS_2025_public.pdfLarge Language Models for Literary Analysis: Detecting Narrative Functions of Parasitic Poems in Late Imperial Chinese Vernaculars (2024-)
Large Language Models for Literary Analysis: Detecting Narrative Functions of Parasitic Poems in Late Imperial Chinese Vernaculars (2024-)
The project is supported by the EMPOWER program at Indiana University Bloomington (2024-2025).
Preprint: https://tuprints.ulb.tu-darmstadt.de/30147/
Abstract: What narrative functions do poems serve when interwoven with vernacular prose? This article takes what has often been labeled as "embedded poems" or "parasitic poems" in late imperial Chinese fiction as the primary subject of study. We examine the narrative roles of these poems within a selected corpus of Qing dynasty fiction, specifically investigating if an approach that combines human annotation with large language models can aptly capture and automatically classify their narrative functions. Through two rounds of iterative annotation and large language model testing, we demonstrate both the potential and limitations of this approach. As one of the few studies that applies large language models to Chinese literary research, our work lays the groundwork for future large-scale investigations into the dynamics between verse and prose in classical Chinese literature, incorporating both canonical works and beyond.
Abstract: What narrative functions do poems serve when interwoven with vernacular prose? This article takes what has often been labeled as "embedded poems" or "parasitic poems" in late imperial Chinese fiction as the primary subject of study. We examine the narrative roles of these poems within a selected corpus of Qing dynasty fiction, specifically investigating if an approach that combines human annotation with large language models can aptly capture and automatically classify their narrative functions. Through two rounds of iterative annotation and large language model testing, we demonstrate both the potential and limitations of this approach. As one of the few studies that applies large language models to Chinese literary research, our work lays the groundwork for future large-scale investigations into the dynamics between verse and prose in classical Chinese literature, incorporating both canonical works and beyond.
Presentation1.pptxBridging Sight and Insight: Visualization in Action among Digital Humanists (Dissertation Project) (2020-2022)
Bridging Sight and Insight: Visualization in Action among Digital Humanists (Dissertation Project) (2020-2022)
My dissertation won the Eugene Garfield Doctoral Dissertation Fellowship (2021) from Beta Phi Mu Honor Society and the Best Doctoral Dissertation Proposal Award (2021) from the Association for Information Science and Technology.
Abstract: Visualization has long been recognized as an essential component of scientific research for knowledge discovery and communication. In his classic essay titled “Drawing Things Together,” Latour (1990) argued that visual inscriptions are the “immutable mobiles” in sciences, which capture the fundamental scientific discovery and information within their standardized, optically consistent structures, while at the same time, mobilize the construction and diffusion of scientific consensus with their ability to move across various contexts, especially from the laboratories to scientific publications. Empirical results from research in science and technology studies (STS) have confirmed this epistemic role of visualization and demonstrated the value of visualization as a critical indicator for the disciplinarity of scientific fields. In contrast to scientific disciplines, the values and functions of visualization in the humanities have yet to be examined empirically, despite the increasing application of visual technologies in humanities research. My dissertation offers the first thorough empirical examination of visualizations in digital humanities (DH) research using large-scale DH journal publication data and semi-structured interviews with DH practitioners, illustrating visualization's rhetorical roles in humanities knowledge communication and its impacts on the evolving dynamics of DH scholarship.
DH2022Presentation.pptxResearch Assistant, University of Pittsburgh Library System (ULS)
Overview: Chinese Village Gazetteer Data Project (CCVGD) creates an open-access database of contemporary Chinese villages based on 3,000+ village gazetteers that contain a wide variety of information regarding villages' demographics, geographies, climates, education, economics, and healthcare. As a research fellow, I collaborated with the student team, librarians, and faculty advisors across departments and schools to design the CCVG database and a web interface, especially engaging in the use of visualizations in the interface.
Digital Collections of Books of Hours Manuscripts & Visualization (2019-2020)
Digital Collections of Books of Hours Manuscripts & Visualization (2019-2020)
Data Science Researcher, Schoenberg Institute for Manuscript Studies, University of Pennsylvania
This project was awarded the 2019 LEADS-4-NDP Fellowship ($5,000) from Drexel University.
Overview: The Book of Hours is a Christian devotional book popular in the Middle Ages. As a Data Science Researcher with the LEADS-4-NDP program, I experimented with digital methods to visualize the digitized collection of Books of Hours manuscripts indexed in BiblioPhilly, particularly the metadata; performed data cleaning, manipulation, and visualization using Tableau and Python; designed and implemented interactive data visualizations for the project.
Page updated
Google Sites
Report abuse