Hazel Kim

I am a DPhil student in computer science at the University of Oxford.

I conduct research in Natural Language Processing and Machine Learning at Oxford Applied and Theoretical Machine Learning Group (OATML). I am grateful to be advised by Yarin Gal and Sebastian Farquhar from OATML Group. I am also fortunate to be advised by Janet Pierrehumbert from OxNLP and Yonatan Belinkov from Technion as a student of the European Laboratory for Learning and Intelligent Systems (ELLIS Society).

Language Acquisition from Limited Sources

Deep learning shapes the world with given data. It works well with quality and abundant data while providing biased perspectives with scarce, poor data. However, quality data is often expensive to collect or private to get access. This challenge has inspired me to study how to acquire language proficiency from limited data.

Causal Inference, not Correlation

Deep learning understands the world with correlations between input data and output predictions. Due to the heavy reliance on patterns of the observed data, it lacks reasoning about cause-and-effect relations that are not explicitly appeared in the text. High correlations do not always uncover the causality of events. Causal inference is an important topic to control language models against biased data.

Interpretable and Controllable Language Models

The predictable behaviors under solid interpretation and trustworthiness would enable models to be more transparent and applicable to the real world. The controllable language models would allow the learning procedure to de-bias the models with minimum human supervision and less computational cost.

My Wonderful Collaborators

I was fortunate to start researching with consistent support from Yosub Han. I got great insights into what and how to research from Seong Joon Oh. I was delighted to work with humble yet astute advice from Sangdoo Yun and lucky to have inspiring discussions with Kangmin Yoo. I was happy to mentor enthusiastic JB. Kim in writing his first paper. I enjoyed working on my first paper with Daecheol Woo's sincerity and positive mindset. I thankfully met many great people while conducting research! I appreciate all of my collaborators for supporting me in many different ways :)

Publications

• Selected and Recent Papers

Resume

• University of Oxford 2023 - now

  Postgraduate
  OATML Group
• CLASSTING AI RESEARCH 2022 - 2023

  AI Research
  Rationales and Reasoning with Large Language Models
• NAVER AI LAB 2021 - 2022

  AI Research
  Commonsense Reasoning
• Yonsei University 2020 - 2022

  Master in Artificial Intelligence
  Learning with Limited Data Using Compositionality in Language
• Kalamazoo College 2014 - 2019

  B.A. in Computer Science and Political Science
  Founder's Scholar
• Seoul National University 2012 - 2014

  Graduate School of Public Health
  Youth Research Team Leader
• Hankuk Academy of Foreign Studies 2011 - 2014

  International Division
  Latin Honors Society

Acknowledgement

This website uses the website design and template by Martin Saveski