Yushi is a DPhil student in Social Data Science at the OII, and an aspiring data scientist. Her research focuses on developing interpretability methods to improve safety mechanisms in large language models and ensure robust harm mitigation in generated outputs.
Yushi holds an MSc in Statistical Science from the University of Oxford and an MSc in Artificial Intelligence from Imperial College London. Before that, she held a BSc in Mathematics, Operation Research, Statistics, and Economics from the University of Warwick. Prior to the DPhil, she interned with UNDP as a data analyst, at Oxford and Imperial College London as a machine learning researcher, and at a start-up as a data scientist.
Yushi’s experience includes a range of data science projects, developing machine learning models to build privacy-preserving recommender systems, track brand sentiment in online news, predict US gun violence diffusion rates, and interpret people’s sentiments to artworks. Her skillsets include Python, R, SQL, deep learning models, and AWS SageMaker.
Large language models; Interpretability; AI safety and alignment.
