I am a Ph.D. student candidate in the Department of Computer Science at Dartmouth College, advised by Prof. Yujun Yan.
My research focuses on the science of Language Models, particularly on understanding their behavior during both training and inference. My work centers on understanding how post-training shapes LLMs and how LLMs work internally at inference, and on leveraging that understanding to develop technical methods that enhance their performance.
Educations
- 2023.09 - present, Ph.D. in Computer Science, Dartmouth College.
- 2019.09 - 2023.06, B.S. in Computer Science, Nankai University.
News
- 2026.09: One paper was accepted to NeurIPS 2026.
- 2026.06: Started an Applied Scientist internship at Amazon in Seattle, WA.
- 2026.04: Our paper on customizing LLMs’ memory by controlling their hidden states was accepted to the ACL main conference as an oral presentation.
- 2025.06: Started a research internship at NEC Laboratories America in Princeton, NJ.
- 2025.05: One paper was accepted to Findings of ACL 2025.
- 2024.09: One paper was accepted to NeurIPS 2024.
Publications
When Form Changes but Logic Doesn’t: Building Logic-invariant LLMs through Structures
Xuyuan Liu, Xinshuai Dong, Elynn Chen, Yujun Yan
NeurIPS 2026
Representation Interventions Enable Lifelong Unstructured Knowledge Control
Xuyuan Liu,Shengyu Chen, Xinshuai Dong, Yanchi Liu, Xujiang Zhao, Haoyu Wang, Yujun Yan, Haifeng Chen,Zhengzhang Chen
Spectral Insights into Data-Oblivious Critical Layers in Large Language Models
Xuyuan Liu, Lei Hsiung, Yaoqing Yang, Yujun Yan
Findings of ACL 2025, also in Unireps@NeurIPS 2025
Exploring Consistency in Graph Representations: from Graph Kernels to Graph Neural Networks
Xuyuan Liu, Yinghao Cai, Qihui Yang, Yujun Yan
NeurIPS 2024
TreeMAN: Tree-enhanced Multimodal Attention Network for ICD Coding
Zichen Liu, Xuyuan Liu, Yanlong Wen, Guoqing Zhao, Fen Xia, Xiaojie Yuan
COLING 2022 (ORAL)
Honors
- 2023, Dartmouth Fellowship
- 2022, Academic Excellence Scholarship, Nankai University
- 2022, Scientific Research Innovation Scholarship, Nankai University
Services
- Conference Reviewer: NeurIPS 2024-2025, ICLR 2025-2026, ICML 2025-2026, KDD Dataset & Benchmark 2026