Minhwa Lee

Full-time Visiting Researcher, University of Minnesota - Twin Cities

minhwalee [AT] umass.edu

Hello!

I am currently a visiting researcher in Minnesota NLP group at University of Minnesota Twin Cities (UMN), advised by Prof. Dongyeop Kang. Previously, I obtained my Bachelor's degrees in Computer Science and Mathematics from The College of Wooster, and Master's degree in Computer Science at University of Massachusetts Amherst.

Broadly, my research interests lie in the intersection of natural language processing (NLP), human-AI interaction, and education. In particular, I aim to develop and evaluate human-AI collaborative systems for educational support (e.g., writing) powered by large language models (LLMs) and interactive design frameworks. Also, I am interested in developing socially-aware natural language processing (NLP) technologies that better aligns with human reasoning and promote diversity and equity in modern NLP technologies. Before joining UMN, I used to conduct research on addressing health disparity issues from social media texts using data mining and NLP approaches.

Note: I am looking for full-time research or job opportunities starting Summer 2024! Here is my full CV.

Updates

  • [May. 2024]    Our preprint about benchmarking LLM evaluations on cognitive biases are accepted to Findings of ACL 2024!
  • [Mar. 2024]    Elated to announce that my first NLP+HCI paper "Human-AI Collaborative Taxonomy Construction: A Case Study of Profession-specific Writing Assistants" is accepted to In2Writing Workshop @ CHI 2024!
  • [Feb. 2024]    Excited to announce that our work "LocalTweet to LocalHealth: A Mental Health Surveillance Framework Based on Twitter Data" is accepted to LREC-COLING 2024!
  • [Jan. 2024]    A new preprint is out! My first 1st-authored paper in NLP domain! Check out how various types of LLM-generated data still lacks "humanness" through several stress testing!
  • [Nov. 2023]    A new preprint is out! Check out how much our novel method of criteria-based prompting can promote diversity and inclusion in LLMs!
  • [Sep. 2023]    A new preprint about LLM evaluation is out! Check out how much LLM evaluation are affected by cognitive biases and compare it with human alignment as well!
  • [June. 2023]    Elated to announce that I will be working as a full-time visiting researcher at Minnesota NLP!

Publications

Most recent publications on Google Scholar.
‡ indicates equal contribution.

Benchmarking Cognitive Biases in Large Language Models as Evaluators

Ryan Koo, Minhwa Lee, Vipul Raheja, Jong Inn Park, Zae Myung Kim, Dongyeop Kang

To Appear at Findings of ACL 2024

Paper Data

Human-AI Collaborative Taxonomy Construction: A Case Study of Profession-specific Writing Assistants

Minhwa Lee, Zae Myung Kim, Vivek A. Khetan, Dongyeop Kang

To Appear at The Third In2Writing Workshop at CHI 2024

Local Tweets to Local Health: A Mental Health Surveillance Framework Based on Twitter Data

Vijeta Deshpande, Minhwa Lee, Zonghai Yao, Zihao Zhang, Jason B. Gibbons, Hong Yu

To Appear at LREC-COLING 2024

Paper

Under the Surface: Tracking the Artifactuality of LLM-Generated Data

Debarati Das‡, Karin De Langis‡, Anna Martin‡, Jaehyung Kim‡, Minhwa Lee‡, Zae Myung Kim‡, Shirley Hayati, Risako Owan, Bin Hu, Ritik Parkar, Ryan Koo, Jonginn Park, Aahan Tyagi, Libby Ferland, Sanjali Roy, Vincent Liu, Dongyeop Kang

Preprint (2024)

Paper Code Data

How Far Can We Extract Diverse Perspectives from Large Language Models? Criteria-Based Diversity Prompting!

Shirley Anugrah Hayati, Minhwa Lee, Dheeraj Rajagopal, Dongyeop Kang

Preprint (2023)

Paper

Vision Meets Definitions: Unsupervised Visual Word Sense Disambiguation Incorporating Gloss Information

Sunjae Kwon, Rishabh Garodia, Minhwa Lee, Zhichao Yang, Hong Yu

The 61st Annual Meeting of the Association for Computational Linguistics (ACL). 2023.

Paper Code

Benchmarking Cognitive Biases in Large Language Models as Evaluators

Ryan Koo, Minhwa Lee, Vipul Raheja, Jong Inn Park, Zae Myung Kim, Dongyeop Kang

To Appear at Findings of ACL 2024

Paper Data

Human-AI Collaborative Taxonomy Construction: A Case Study of Profession-specific Writing Assistants

Minhwa Lee, Zae Myung Kim, Vivek A. Khetan, Dongyeop Kang

To Appear at The Third In2Writing Workshop at CHI 2024

Local Tweets to Local Health: A Mental Health Surveillance Framework Based on Twitter Data

Vijeta Deshpande, Minhwa Lee, Zonghai Yao, Zihao Zhang, Jason B. Gibbons, Hong Yu

To Appear at LREC-COLING 2024

Paper

Under the Surface: Tracking the Artifactuality of LLM-Generated Data

Debarati Das‡, Karin De Langis‡, Anna Martin‡, Jaehyung Kim‡, Minhwa Lee‡, Zae Myung Kim‡, Shirley Hayati, Risako Owan, Bin Hu, Ritik Parkar, Ryan Koo, Jonginn Park, Aahan Tyagi, Libby Ferland, Sanjali Roy, Vincent Liu, Dongyeop Kang

Preprint (2024)

Paper Code Data

Socioeconomic and Geographic Disparities... [Full Title to be disclosed after acceptance]

Minhwa Lee‡, Yifan Zhang‡, Jason B. Gibbons, Huan-Yuan Chen, Zonghai Yao, Yan Wang, Feiyun Ouyang, David Levy, Katherine L. Tucker, Hong Yu

Under Review by Nature Food, 2024

How Far Can We Extract Diverse Perspectives from Large Language Models? Criteria-Based Diversity Prompting!

Shirley Anugrah Hayati, Minhwa Lee, Dheeraj Rajagopal, Dongyeop Kang

Preprint (2023)

Paper

Vision Meets Definitions: Unsupervised Visual Word Sense Disambiguation Incorporating Gloss Information

Sunjae Kwon, Rishabh Garodia, Minhwa Lee, Zhichao Yang, Hong Yu

The 61st Annual Meeting of the Association for Computational Linguistics (ACL). 2023.

Paper Code

Statistical and Machine Learning Approaches to Depressive Disorders Among Adults in the United States: From Factor Discovery to Prediction Evaluation

Minhwa Lee

Honors Thesis at The College of Wooster, 2021.

Paper Poster

Vitæ

This website was built based on a template by Prof. Martin Saveski. Thank you for sharing!
Minhwa Lee
minhwalee [AT] umass.edu