Ph.D. Student, Computer Science
The University of Texas at Austin
xiye [at] cs.utexas.edu xiye17 xiye_nlp
I am a fourth-year Ph.D. student in the Department of Computer Science at UT Austin, advised by Greg Durrett. I am also a member of the TAUR lab (Text Analysis, Understanding, and Reasoning).
My research focuses on Natural Language Processing and the interpretability of NLP models in particular. Specifically, I am interested in 1) using explanations to improve model performance in downstream applications, and 2) building and evaluating explanations for models designed for complex NLP tasks. I also works on semantic parsing and program synthesis.
Prior to joining UT Austin, I obtained my Bachelor's degree from the School of Software, Tsinghua University, where I worked with Prof. Shixia Liu on Visual Analysis and Machine Learning.
(preprint) The Unreliability of Explanations in Few-Shot In-Context Learning
ArXiv Preprint, 2022
Can Explanations Be Useful for Calibrating Black Box Models? code
Annual Meeting of the Association for Computational Linguistics (ACL), 2022
RnG-KBQA: Generation Augmented Iterative Ranking for Knowledge Base Question Answering code
Annual Meeting of the Association for Computational Linguistics (ACL), 2022
Diagnosing Ensemble Few-Shot Classifiers
Transactions on Visualization and Computer Graphics (TVCG), 2022
Connecting Attributions and QA Model Behavior on Realistic Counterfactuals code
Conference on Empirical Methods in Natural Language Processing (EMNLP), 2021
Optimal Neural Program Synthesis from Multimodal Specifications code
Fingdings of the Conference on Empirical Methods in Natural Language Processing (Findings of EMNLP), 2021
Benchmarking Multimodal Regex Synthesis with Complex Structures code data
Annual Meeting of the Association for Computational Linguistics (ACL), 2020
Sketch-Driven Regular Expression Generation from Natural Language and Examples code
Transactions of the Association for Computational Linguistics (TACL), 2020
Multi-Modal Synthesis of Regular Expressions code
Conference on Programming Language Design and Implementation (PLDI), 2020
Interactive Correction of Mislabeled Training Data video
Conference on Visual Analytics Science and Technology (VAST), 2019
torchASN: a pytorch implementation of Abstract Syntax Network. code
Reviwer: NAACL (22), CONLL (20, 21), SUKI Workshop (22), NLP4Prog Workshop (21).
TA for CS388: Natural Language Processing.
TA for CS429: Computer Architecture and Organization.