Scholar
Junjielong Xu
Google Scholar ID: e-zP90wAAAAJ
The Chinese University of Hong Kong, Shenzhen
Citations & Impact
Citations
220
H-index
5
i10-index
4
Publications
6
Co-authors
5
list available
Publications
Resume (English only)
Academic Achievements
- Publications:
- - AL-Bench: A Benchmark for Automatic Logging (arXiv)
- - Repo2Run: Automated Building Executable Environment for Code Repository at Scale (NeurIPS’25)
- - OpenRCA: Can Large Language Models Locate the Root Cause of Software Failures? (ICLR’25)
- - Aligning the Objective of LLM-based Program Repair (ICSE’25)
- - DivLog: Log Parsing with Prompt Enhanced In-Context Learning (ICSE’24)
- - UniLog: Automatic Logging via LLM and In-Context Learning (ICSE’24)
- - Hue: A User-Adaptive Parser for Hybrid Logs (FSE’23)
- Awards:
- - Duan Yong Ping Travel Award, CUHK-Shenzhen (¥10,000), 2024.04
- - 3rd Prize, CCF International AIOps Challenge (Ranked 1st among student teams), 2023.12
- - 1st Prize, China Undergraduate Mathematical Contest in Modeling (CUMCM) (Leader), 2020.10
- - Finalist Prize, Mathematical Contest In Modeling (MCM) (Solo), 2020.04
Research Experience
- Currently a visiting scholar in AST Lab, ETH Zurich, Mentor: Zhendong Su; Internship at ByteDance.
Education
- Ph.D. candidate, School of Data Science, The Chinese University of Hong Kong, Shenzhen, Supervisor: Pinjia He; BEng, School of Electronic Information and Communication, Huazhong University of Science and Technology, 2022.
Background
- Research interests: LLM-based DevOps automation for enhancing reliability of code and services; other research topics include scalable oversight, reinforcement learning, and test-time scaling. Objective is to provide non-trivial insights into AI models to address impactful challenges in real-world SE scenarios.
Miscellany
- My name “Junjielong Xu” is written as “徐俊杰龙” in Chinese, and is pronounced like “Joon-jyeh-long Shoo.”
