Scholar
Xinyu Li
Google Scholar ID: gIFmXNkAAAAJ
Carnegie Mellon University
Citations & Impact
Citations
287
H-index
4
i10-index
2
Publications
8
Co-authors
0
Publications
Browse publications on Google Scholar (top-right) ↗
Resume (English only)
Academic Achievements
- Publications: 'TimeSeriesGym: A Scalable Benchmark for (Time Series) Machine Learning Engineering Agents', arXiv preprint arXiv:2505.13291, 2025; 'Personalized language modeling from personalized human feedback', NeurIPS 2024 Workshop on Adaptive Foundation Models, 2024.
Research Experience
- Started an Applied Scientist internship at Amazon in Palo Alto in May 2025; Interned at the Biomedical Imaging team at Microsoft Research Cambridge in June 2024; Involved in multiple research projects, including TimeSeriesGym and Personalized RLHF.
Education
- Ph.D.: School of Computer Science at Carnegie Mellon University, Auton Lab, advised by Prof. Artur Dubrawski; MISM: Master of Information Systems Management from Carnegie Mellon University; B.S.: Computer Science (Information Security) from Shanghai Jiao Tong University, China.
Background
- Research Interests: LLM-based agents, particularly for machine learning (ML) engineering tasks. Recent work includes: Benchmarking LLM agents on ML engineering tasks (TimeSeriesGym), developing agents that automate end-to-end ML engineering workflows, improving LLMs for agentic use through multi-tasking and personalization (P-RLHF). Professional field: Computer Science, with a focus on Information Security Management.
Miscellany
- Feel free to reach out to me if you are interested in my research and would like to collaborate!
Co-authors