Unsupervised Evaluation of Code LLMs with Round-Trip Correctness
International Conference on Machine Learning (ICML), 2024
About Me
I am currently a research scientist at Google DeepMind, working on Gemini post-training for agentic coding. Previously, I completed my PhD in Computer Science at The University of Texas at Austin in 2022. I was advised by Prof. Raymond Mooney and Prof. Junyi Jessy Li. My thesis work focused on building AI models for streamlining software engineering. I was a recipient of the 2020-2021 Bloomberg Data Science Ph.D. Fellowship. Before that, I completed my B.S. in Computer Science at the University of Illinois at Urbana-Champaign in 2016.
Publications
Using Developer Discussions to Guide Fixing Bugs in Software
Findings of Conference on Empirical Methods in Natural Language Processing (EMNLP), 2022
Facilitating Software Evolution through Natural Language Comments and Dialogue
PhD Thesis, 2022
CoditT5: Pretraining for Source Code and Natural Language Editing
International Conference on Automated Software Engineering (ASE), 2022
Updated Headline Generation: Creating Updated Summaries for Evolving News Stories
Association for Computational Linguistics (ACL), 2022
Learning to Describe Solutions for Bug Reports Based on Developer Discussions
Findings of Association for Computational Linguistics (ACL), 2022
Deep Just-In-Time Inconsistency Detection Between Comments and Source Code
The AAAI Conference on Artificial Intelligence (AAAI), 2021
Copy That! Editing Sequences by Copying Spans
The AAAI Conference on Artificial Intelligence (AAAI), 2021
Leveraging Class Hierarchy for Code Comprehension
NeurIPS Workshop on Computer-Assisted Programming, 2020
Learning to Update Natural Language Comments Based on Code Changes
Association for Computational Linguistics (ACL), 2020
Associating Natural Language Comment and Source Code Entities
The AAAI Conference on Artificial Intelligence (AAAI), 2020