☕️
About Me
Hi there! I’m TANG Xuemei — or just call me Xuemei. 👋
I earned my PhD from Peking University in 2024, under the supervision of Prof. Jun Wang and Prof. Qi Su, and I’m currently a postdoctoral researcher at The Hong Kong Polytechnic University, working with Prof. Chu-Ren Huang and Dr. Jinghang Gu.
My research lies at the intersection of Natural Language Processing (NLP) and Digital Humanities, where I explore how computational models can help us better understand language, culture, and history.
I’m currently seeking postdoctoral or assistant professor opportunities for next year.
Interests
- Evaluation of Large Language Models (LLMs)
- Information Extraction
- Digital Humanities
- Chinese Cultural Studies
Education
-
PhD Information Science
Peking University
-
MEng Comutational Linguistics
Beijing Normal University
-
BSc Information Engineering
Tianjin Normal University
📚 My Research
Over the past few years, my research has primarily focused on classical Chinese historical documents. I aim to leverage natural language processing (NLP) to develop foundational tools for historical document studies and to support content-oriented research. I am also interested in evaluating large language models (LLMs). Currently, I am continuing to explore research in cultural heritage, including multimodal directions.
Please feel free to reach out if you are interested in collaboration.😃
Featured Publications
CHisIEC: An Information Extraction Corpus for Ancient Chinese History
Recent Publications
(2025).
Joint Information Extraction Across Classical and Modern Chinese with Tea-MOELoRA.
arXiv:2509.01158.
(2025).
Large Language Models for Automated Literature Review: An Evaluation of Reference Generation, Abstract Writing, and Review Composition.
In EMNLP 2025 Main.
(2025).
Language model collaboration for relation extraction from classical Chinese historical documents.
Information Processing & Management.
(2025).
ChisNERE: a premodern Chinese corpus with named entity and relation annotation.
Digital Scholarship in the Humanities.
(2025).
An Effective Incorporating Heterogeneous Knowledge Curriculum Learning for Sequence Labeling.
In ACL 2025 Main (Oral).
Recent & Upcoming Talks
Recent News