Ruiqi Zhong (ruiqi-zhong@berkeley.edu)

I am currently a second year PhD student in the Berekely NLP group, advised by Prof. Dan Klein and Prof. Jacob Steinhardt. I am broadly interested in the intersection of NLP with programming languages and machine learning. Specifically, I work on

  • Text-to-progrm generation: how do we map human languages to executable programs, in an interactive and multi-modal (e.g programming by examples) way.

  • Deep learning training and generalization: can we explain and predict what and how neural networks learn?
  • Stylistically, I prefer projects that unlock new lines of researches, or rethink the current conceptual framework.

    I finished my undergrad at Columbia University and was advised by Prof. Kathleen McKeown.

    Twitter Google Scholar GitHub

    Mentorship

    Ph.D. students from the Berkeley NLP group have compiled a list of resources to prepare Berkeley undergrads for NLP researches. I am also happy to talk about research personally. Just send me an email if you wanna chat with me.

    If you specifically want to work with me, try finishing the project 2 of CS288 before contacting me.

    Publications

    Meta-tuning Language Models to Answer Prompts Better
    Ruiqi Zhong, Kristy Lee, Zheng Zhang, Dan Klein
    arXiv, 2021
    [paper]

    Approximating How Single Head Attention Learns
    Charlie Snell, Ruiqi Zhong, Jacob Steinhardt, Dan Klein
    arXiv, 2021
    [paper] [slides] [code] [blog]

    Semantic Evaluation for Text-to-SQL with Distilled Test Suites
    Ruiqi Zhong, Tao Yu, Dan Klein
    EMNLP 2020
    [paper] [code] [slides]

    Semantic Scaffolds for Pseudocode-to-Code Generation
    Ruiqi Zhong, Mitchell Stern, Dan Klein
    ACL 2020
    [paper] [code] [video] [slides]

    Detecting and Reducing Bias in a High Stakes Domain
    Ruiqi Zhong, Yanda Chen, Desmond Patton, Charlotte Selous, Kathleen McKeown
    EMNLP 2019
    [paper] [code] [poster]

    Fine-grained Sentiment Analysis with Faithful Attention
    Ruiqi Zhong, Steven Shao, Kathleen McKeown
    arXiv preprint
    [paper]

    Subspace Embedding and Linear Regression with Orlicz Norm
    Alexandr Andoni, Chengyu Lin, Ying Sheng, Peilin Zhong, Ruiqi Zhong
    ICML 2018
    [paper] [video] [slides]

    Detecting Gang-Involved Escalation on Social Media using Context
    Serina Chang, Ruiqi Zhong, Ethan Adams, Fei-Tzin Lee, Siddharth Varia, Desmond Patton, William Frey, Chris Kedzie, Kathleen McKeown
    EMNLP 2018
    [paper] [code] [video]

    Paper Reading Guides (Chinese)

    Debiasing NLP: Man is to Computer Programmer as Woman is to Homemaker?
    Ruiqi Zhong, Yanda Chen, Junyao Shi

    ICML2018 Fairness ML Best Papers.
    Ruiqi Zhong, Xinyi Han

    Miscellaneous

    I represented Columbia University in ACM-ICPC and Putnam Math Competition during my Sophormore year (though it seems I was the bottle-neck of our teams).
    I sleep at 11 p.m. and do not respond to later messages. Sometimes, however, I am actually awake; but I pretend not to see them anyways.
    My favorite animation character and role model is Wenli Yang in Legend of Galactical Heroes.