My research happens in a network of collaborations across educational measurement, computational social science, and computational genomics. This page introduces the people I work with — and leaves room for the team to grow.
My work currently spans four complementary directions, each supported by a different group:
Educational measurement & NLP — with Prof. Hong Jiao’s group at the University of Maryland (since Dec. 2022), I work on automated essay scoring, LLM-based scoring rationales, and generative-AI text detection for K–12 and AP assessment contexts, with collaborators at the University of Iowa and Beijing Normal University.
Computational social science — at Princeton University, I support interdisciplinary research using machine learning and multimodal data to study human behavior and communication on social media.
Computational genomics — at Westlake University, I work at the intersection of computational genomics, machine learning, and deep learning, with an emphasis on genomic sequence modeling, molecular phenotype prediction, model generalization, and rigorous benchmarking.
Computational neuroscience — my undergraduate focus at Princeton, where I study how computational models can help us understand complex biological and cognitive systems.
I am actively looking to connect with new labs and mentors in neuroscience, genomics, AI for science, and interdisciplinary computation.
Leads the collaboration on automated essay scoring, LLM scoring rationales, and generative-AI text detection.
Co-author on LLM scoring rationale comparisons and automated essay scoring in long contexts.
Co-author on generative-language-model summarization for scoring long essays.
Co-author on AI-generated text detection — Frontiers in AI 2024 and IMPS 2024 proceedings.
Co-author on automated essay scoring in long contexts (NCME 2025 eBoard).
Co-author on exploring the utilities of LLM rationales for automated essay scoring.
Professor whose freshman seminar Evolution of Human Language I supported as teaching assistant and official note taker.
Slot reserved for future advisors, lab members, and collaborators — including a Princeton neuroscience mentor.
LLM rationales, summarization-based scoring, and empirical comparisons — with Hong Jiao (UMD), Dan Song (Iowa), Xinyi Wang, Kuo Wang, and others.
Tokenization, dataset size, and revision conditions in identifying AI-generated text — with Jiayu Yao.
Genomic sequence modeling, molecular phenotype prediction, and rigorous benchmarking with deep-learning pipelines at scale.
Multimodal machine learning on human behavior and communication in online discourse and social interaction.
Upcoming collaborations on neuroscience and behavioral data — placeholder for active projects.