Education
- Ph.D. in Statistics (dual program with Computer Science), Michigan State University, East Lansing, USA — Sept 2023 – Present
- M.S. in Statistics, University of Washington, Seattle, USA — Sept 2021 – Mar 2023
- B.S. in Computing Mathematics (Minor: Computing), City University of Hong Kong, Hong Kong — Sept 2017 – Jun 2021
Research experience
- Research Assistant — Michigan State University — Dec 2024 – Present
- Developing an AI auto-grading system; deploying/fine-tuning LLMs and adding inference-time methods to improve rubric-aligned accuracy and reasoning.
- Teaching Assistant (STT 200) — Michigan State University — Sept 2023 – Present
- Lead discussion/lab sections, office hours, and assessment support; build rubrics and checks for introductory statistics.
- Academic Student Employee (Research Assistant) — University of Washington — Jun 2022 – Sept 2022
- Advisors: Prof. Elena Erosheva, Prof. Marco Corneli.
- Extended a Bayesian mixture model for longitudinal trajectories to a heterogeneous model; implemented imputation for Alzheimer’s diagnostics; contributed to manuscripts.
- Research Assistant (Deep Learning Optimization) — City University of Hong Kong — Jun 2020 – Jun 2021
- Advisor: Prof. Daniel W. C. Ho.
- Implemented SGD and a CNN pipeline on CIFAR-10 (training accuracy >90%); studied over/under-sampling on imbalanced data.
Publications
Statistics and Computer Science Theory
- Corneli, M., Erosheva, E., Qian, X., et al. (2025). A Bayesian approach for clustering and exact finite-sample model selection in longitudinal data mixtures. Computational Statistics, 40, 509–545. https://doi.org/10.1007/s00180-024-01501-5
- Qian, X., & Xing, Y. Ensuring Calibration Robustness in Split Conformal Prediction Under Adversarial Attacks. Submitted to ICML 2026.
- Qian, X., Cui, Y., Lin, Y., Wu, J., & Xing, Y. Multi-source online conformalized jailbreak detection. Submitted to UAI 2026.
Education
- Shin, N., Xing, Y., Qian, X., & Krajcik, J. (2025). Developing a Generative AI Framework for Analyzing Student Responses to Enhance Classroom Assessments. In ICLS 2025 (pp. 2711–2713).
- Shin, N., Qian, X., Chu, Y., Li, H., Miller, C., Krajcik, J., & Xing, Y. AI in assessment. Paper presented at the Global Initiative in AI and Emerging Technologies in STEM Education. 2025.
- Shin, N., Qian, X., Chu, Y., Li, H., Miller, C., Krajcik, J., & Xing, Y. Multi-agent systems for detecting uncertainty and weaknesses in elementary students’ written responses to usable knowledge tasks. Paper presented at the Georgia Conference on AI and Education. 2025.
- Qian, X., Shin, N., Li, H., Chu, Y., Miller, C., Krajcik, J., Tang, J., & Xing, Y. Multi-agent Large Language Model systems for Analyzing Elementary Students’ Constructed Response. Accepted to AERA 2026.
- Shin, N., Qian, X., Li, H., Chu, Y., Miller, C., Tang, J., Krajcik, J., & Xing, Y. An AI framework for identifying uncertainty and weaknesses in written responses to usable knowledge tasks. Paper to be presented at the 2026 AERA Annual Meeting.
- Shin, N., Qian, X., Li, H., Chu, Y., Miller, C., Tang, J., Krajcik, J., & Xing, Y. Enhancing scoring accuracy with multi-agent systems: Analyzing elementary students’ written responses in science assessment. Poster to be presented at the 2026 NCME Annual Meeting.
- Shin, N., Qian, X., Li, H., Chu, Y., Miller, C., Tang, J., Krajcik, J., & Xing, Y. Leveraging generative AI to detect uncertainty in elementary students’ written science responses. Paper to be presented at the 2026 NARST Annual Meeting.
- Shin, N., Qian, X., Li, H., Chu, Y., Miller, C., Krajcik, J., Tang, J., & Xing, Y. A Generative AI Framework for Analyzing Student Responses in Formative Assessments to Provide Targeted Feedback. Submitted.
