Xiaoyu Liu @ Microsoft

Biography

I am currently a Data & Applied Scientist at Microsoft Visual Studio and GitHub team. I have graduated as a PhD at Southern Methodist University Department of Computer Science. I was advised by Prof. LiGuo Huang. I also worked with collegues from Human Language Technology Research Institute in UTD to apply machine learning on software engineering.

I am currently interestd in solving software engineering tasks using large-scale learning-based language model. I worked on a lot of research topics such as code generation, API recommendation, software licensing, software change impact analysis, and software change intent analysis. Details can be found in my publications.

News

Our “Learning to Reduce False Positives in Analytic Bug Detectors” paper has been accepted by ICSE 2022!

Publications

Learning to Reduce False Positives in Analytic Bug Detectors

Anant Kharkar, Roshanak Zilouchian Moghaddam, Matthew Jin, Xiaoyu Liu, Xin Shi, Colin Clement, Neel Sundaresan.
The 44th International Conference on Software Engineering (ICSE 2022).

Long-Range Modeling of Source Code Files with eWASH: Extended Window Access by Syntax Hierarchy

Colin B Clement, Shuai Lu, Xiaoyu Liu, Michele Tufano, Dawn Drain, Nan Duan, Neel Sundaresan, Alexey Svyatkovskiy.
Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing (EMNLP 2021).

Predicting Licenses for Changed Source Code

Xiaoyu Liu, LiGuo Huang, Jidong Ge, Vicent Ng.
The 34th IEEE/ACM International Conference on Automated Software Engineering (ASE 2019).

Effective API Recommendation without Historical Software Repositories

Xiaoyu Liu, LiGuo Huang, Vicent Ng.
The 33rd IEEE/ACM International Conference on Automated Software Engineering (ASE 2018).

Linking Source Code to Untangled Change Intents

Xiaoyu Liu, LiGuo Huang, Chuanyi Li, Vincent Ng.
The 34th IEEE International Conference on Software Maintenance and Evolution (ICSME 2018).

Do code data sharing dependencies support an early prediction of software actual change impact set?

Xiaoyu Liu, LiGuo Huang, Alexander Egyed, Jidong Ge.
Journal of Software: Evolution and Process, 2018.

Patents

US Patent 11,243,750: Code completion with machine learning
US Patent 11,175,897: Language interoperability to automate code analysis
US Patent App. 16/900,993: Neural Code Completion via Re-ranking

Recent Services

IEEE Transactions on Software Engineering (TSE) Co-Reviewer
International Conference on Software and System Processes (ICSSP) 2017, 2018 Co-Reviewer
IEEE International Conference on Information Reuse and Integration (IRI 2018) 2018 Co-Reviewer
Asia-Pacific Software Engineering Conference (APSEC) 2017, 2018, 2019 Co-Reviewer

XIAOYU LIU

Biography

News

Publications

Patents

Recent Services

Connections