Yulin Chen
New York, New York, United States
Summary
🤩
Rockstar🎓
Top SchoolYulin Chen is a first-year PhD student in Data Science at NYU with six years of research and engineering experience focused on NLP and large language model behavior. Previously a research assistant at Tsinghua, Yulin worked on LLM alignment, efficient prompting methods, and constructing human-preference alignment datasets while exploring ways to merge model abilities. They contributed to the popular OpenPrompt open-source framework, improving prompt template generation and adding memory-control features to make prompt learning more resource-efficient. Yulin blends computational linguistics (BA in English/Linguistics) with engineering rigor (MEng in Computer Engineering), bringing a rare cross-disciplinary perspective to problems in model interpretability and alignment. Their current research probes how LLM internals map to linguistic theory and how those insights can directly improve model design. Colleagues describe them as a practical researcher who bridges theory, tooling, and dataset engineering to push LLMs toward safer, more explainable behavior.
6 years of coding experience
3 years of employment as a software developer
高中, 高中 at Hangzhou Foreign Languages SchoolDoctor of Philosophy - PhD Data Science, Doctor of Philosophy - PhD Data Science at New York UniversityMaster of Engineering - MEng Computer Engineering, Master of Engineering - MEng Computer Engineering at Tsinghua University
Github Skills (9)
pytorch10
transformer10
nlp10
deep-learning10
python10
natural-language-processing9
ai9
nlpjs8
natural-language-understanding8
Programming languages (2)
Github contributions (5)
thunlp/OpenPrompt
Oct 2021 - Apr 2022
An Open-Source Framework for Prompt-Learning.
Role in this project:
ML Engineer
ML EngineerContributions:39 commits, 9 PRs, 21 pushes in 6 months
Contributions summary:Yulin contributed to the development of an open-source framework for prompt-learning. Their work involved modifying the `prompt_generator.py` file, which suggests an involvement in the core logic for generating prompts. The commits indicate a focus on template generation, which is a key aspect of prompt-based learning. The changes also incorporate functionalities like memory control, which suggests that the contributions are centered around efficient resource utilization.
natural-language-understandingpre-trained-language-modelsdialogue-systemsnatural-language-processingprompts
cyl628/save-website-as-pdf
Nov 2020 - Mar 2022
Contributions:2 releases, 17 pushes, 1 branch in 1 year 4 months
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.