About Me

I am a Postdoctoral Researcher in the Center for Language and Speech Processing @ Johns Hopkins University, working with Dr. Mark Dredze. My research interests lie in Trustworthy AI, Natural Language Processing (NLP), and Interpretable Machine Learning. I aim to develop explainable AI techniques that are easily accessible to system developers and end users for building trustworthy and reliable intelligent systems. My current research is centered around trustworthy NLP, with a focus on interpretability, robustness, and fairness, to support the understanding and interplay between humans and neural language models. I obtained my Ph.D. in Computer Science at the University of Virginia, under the supervision of Dr. Yangfeng Ji.

⭐ I will join the Department of Computer Science @ Rice University as a tenure-track Assistant Professor starting Summer 2024. I'm actively looking for motivated students to join my group!

📌 Research Overview

How to understand black-box models?
I develop explanation methods to explain model predictions (ACL'2020, NAACL'2021) and uncertainty (AAAI-UDM'2023) and evaluate explanations to understand model reasoning (ACL'2023) and robustness (BlackboxNLP'2022, 2021)
How to build trustworthy models?
To cultivate trustworthiness in existing models, I have designed data augmentation (NeurIPS WS'2019), variational word masks (EMNLP'2020), and interaction graphs (AAAI'2023), to make model decision-making transparent and reliable
How to improve models via explanations?
I diagnose and debug models, especially their robustness and fairness, through the lens of explanations and develop solutions to improve them (AAAI'2022, ACL WS'2022)

🥳 Received the Outstanding Doctoral Student Award, UVA, 2023

🎉 Received the John A. Stankovic Graduate Research Award, UVA, 2023

🎉 I was awarded the Carlos and Esther Farrar Fellowship, 2022 - 2023

👩‍🏫 As the primary instructor, I co-designed and taught the course, CS 6501/4501 Interpretable Machine Learning, at UVA in Spring 2022, and was awarded the UVA CS Outstanding Graduate Teaching Award and University-wide Graduate Teaching Awards Nominee (top 5% of graduate instructors)

📝 My collaborators and I are actively updating a Reading List with interesting papers

Research Experience

Allen Institute for AI (AI2), Seattle, WA, May 2022 - Oct. 2022
Mosaic Group

Research Intern
Manager: Yejin Choi
Mentors: Swabha Swayamdipta, Faeze Brahman, Xiang Ren
Microsoft Research, Redmond, WA, May 2021 - Aug. 2021
Language and Information Technologies Group
- Research Intern
- Manager: Ahmed H. Awadallah
- Mentors: Guoqing Zheng, Srinagesh Sharma
IBM Research, New York, NY, Jun. 2020 - Aug. 2020
Thomas J. Watson Research Center
- Research Intern
- Manager: Luis Lastras
- Mentors: Chulaka Gunasekara, Song Feng, Hui Wan, Jatin Ganhotra, Sachindra Joshi

Academic Activities

Invited Talk on Bridging the Trustworthy Gap between AI and Humans: Interpretation Techniques for Modern NLP at the CLSP Seminar @ Johns Hopkins University, Mar. 2023
Presentation on Information-Theoretic Evaluation of Free-Text Rationales with Conditional V-Information at Trustworthy and Socially Responsible Machine Learning (TSRML) Workshop @ NeurIPS, Dec. 2022
Presentation on Explaining Predictive Uncertainty by Looking Back at Model Explanations at WiML Workshop 2022 @ NeurIPS, Nov. 2022
Talk on REV: Information-Theoretic Evaluation of Free-Text Rationales @ Allen Institute for AI (AI2), Oct. 2022
Paper presentation on Pathologies of Pre-trained Language Models in Few-shot Fine-tuning Insights Workshop @ ACL 2022, May 2022
Paper presentation on Adversarial Training for Improving Model Robustness? Look at Both Prediction and Interpretation @ AAAI 2022, Feb. 2022
Invited talk on Improving Model Robustness via Interpretation-based Adversarial Training @ MLNLP, Dec. 2021
Presentation on Adversarial Training for Improving Model Robustness? Look at Both Prediction and Interpretation at WiML Workshop 2021 @ NeurIPS, Dec. 2021
Presentation on Adversarial Training for Improving Model Robustness? Look at Both Prediction and Interpretation @ 2021 Fall UVA CS Research Symposium, Dec. 2021
Paper presentation on Explaining Neural Network Predictions on Sentence Pairs via Learning Word-Group Masks @ NAACL 2021, Jun. 2021
2021 CRA-WP Grad Cohort for Women Workshop, Apr. 2021
Poster presentation @ ACM Capital Region Celebration of Women in Computing (CAPWIC), Mar. 2021
Paper presentation on Learning Variational Word Masks to Improve the Interpretability of Neural Text Classifiers @ EMNLP 2020, Nov. 2020
Presentation on Learning Variational Masks for Explainable Next Utterance Prediction in Dialog Systems at IBM Research, Aug. 2020
Paper presentation on Generating Hierarchical Explanations on Text Classification via Feature Interaction Detection @ ACL 2020, Jul. 2020
Poster presentation on Improving the Explainability of Neural Sentiment Classifiers via Data Augmentation @ NeurIPS 2019 Workshop on Robust AI in Financial Services, Dec. 2019
Poster presentation on Building Hierarchical Interpretations in Natural Language via Feature Interaction Detection @ UVA CS Research Symposium Fall 2019, Oct. 2019
Invited talk on How to Train a More Interpretable Neural Text Classifier, AIML-Seminar @ UVA, Apr. 2019
Poster presentation on An Empirical Comparison on Convolutional and Recurrent Neural Networks for NLP at the JUMP Undergraduate Research Initiative, UVA, Nov. 2018

Professional Service

Diversity Representative for UVA Computer Science Graduate Student Group (CSGSG) Council, 2022
Area Chair for WiML Workshop @ NeurIPS 2022
Program Committee: ACL 2023, AAAI 2023, EMNLP 2021 - 2022, NAACL 2021, EACL 2023, CoNLL 2021 - 2022, NLPCC 2022, ACL DialDoc Workshop 2022, EMNLP BlackboxNLP Workshop 2021, 2023, NeurIPS Explainable AI Approaches for Debugging and Diagnosis Workshop 2021, Document-grounded Dialogue Workshop 2021, MASC-SLL 2020
Reviewer: NeurIPS 2023, ACL Rolling Review 2021 - 2022, ACL 2020 - 2021, EMNLP BlackboxNLP Workshop 2022, CoNLL 2019 - 2020, NLPCC 2019 - 2021

Last update: 06/2023