Yangning Li
Web Developer at Tsinghua University
California, United States
Summary
🤩
Rockstar🎓
Top SchoolYangning Li is a web developer and PhD candidate at Tsinghua University specializing in Knowledge-Enhanced Large Language Models, with research emphasis on RAG, continual pre-training, and context compression. Currently a visiting researcher at University of Illinois Chicago with Prof. Philip S. Yu, Yangning combines nine years of engineering experience with hands-on research internships at major labs including Alibaba Cloud and Tencent. He contributes to influential open-source NLP tooling—such as implementing data processors and prompt templates for the popular OpenPrompt framework—demonstrating a focus on practical, production-ready ML components. Based in California with a computer science bachelor’s from Huazhong University of Science and Technology, he favors the philosophy “build something that works,” balancing rigorous academic methods with pragmatic engineering. Notably, his work bridges model-level innovations and engineering plumbing, making research ideas more deployable in real-world NLP tasks.
9 years of coding experience
University of Illinois Chicago学士, 计算机, 学士, 计算机 at 华中科技大学博士, 博士 at 清华
Github Skills (8)
transformer10
pytorch10
nlp10
python10
natural-language-processing10
data-processing10
ai9
machine-learning9
Programming languages (4)
Github contributions (5)
thunlp/OpenPrompt
Sep 2021 - Nov 2022
An Open-Source Framework for Prompt-Learning.
Role in this project:
ML Engineer
ML EngineerContributions:2 releases, 1 review, 92 commits in 1 year 1 month
Contributions summary:Yangning contributed to the development of the `openprompt` framework, focusing on data processing and template implementations for prompt learning. Their work included the creation of data processors for various FewGLUE tasks, such as RTE, CB, WiC, WSC, BoolQ, and COPA, as well as modifications to existing pipeline and prompt base functionalities. The commits demonstrate a focus on integrating and utilizing pre-trained language models for natural language processing tasks.
natural-language-understandingpre-trained-language-modelsdialogue-systemsnatural-language-processingprompts
thunlp/Few-NERD
May 2021 - Oct 2022
Code and data of ACL 2021 paper "Few-NERD: A Few-shot Named Entity Recognition Dataset"
Contributions:44 commits, 3 PRs, 24 pushes in 1 year 5 months
nlpentity-typingnamed-entity-recognitionentity-recognitiondeep-learning
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.