Huawei 2012 Theory Lab
Researcher
Yuqi Zuo
About
I am a researcher interested in foundation models, with a focus on the data and systems that make language models more capable, efficient, and reliable.
My current work spans post-training data, pre-training data laws, retrieval-augmented generation, and efficient language models. I am especially interested in turning empirical observations into practical methods that scale.
You can reach me at zyqdzkd@outlook.com, or find me on GitHub and Google Scholar.
Experience
Microsoft AI & Microsoft Research Asia
Research intern · Information retrieval for ads and language models
Research
Data
Data curation, scaling laws, and post-training methods for capable and reliable foundation models.
Efficiency
Efficient inference and deployment for large language models, with a focus on MoE systems.
Retrieval
Large-scale retrieval and recommendation using semantic representations and generative methods.
Publications
Publications are being updated. For the latest work or collaboration enquiries, please get in touch.