Faculty
Email：xiaodong.gu@sjtu.edu.cn
Institute：Institute of Scalable Computing
MainPage：https://guxd.github.io
Brief Introduction
Xiaodong Gu is an Associate Professor at the School of Computer Science, Shanghai Jiao Tong University. His primary research interests include large language models, intelligent software engineering, and natural language processing, focusing on developing efficient machine learning algorithms for software code. His research topics encompass code large language models, program generation and repair, and intelligent agent-based question answering. His research findings have been published in prestigious international journals and conferences such as ICSE, ICLR, FSE, AAAI, ASE, and ICPC. He is the lead instructor for foundational artificial intelligence courses at the School of Computer Science, including "Machine Learning" and "Fundamentals of Computational Mathematics." He has presided over and participated in more than ten research projects, including those funded by the National Natural Science Foundation of China, the National Key Research and Development Program of China, the Shanghai Natural Science Foundation, as well as corporate projects with companies like Huawei and Tencent. He has been honored with the Shanghai Overseas High-Level Talent Program and the Huawei Spark Award.
Teaching Assignment
SE3332 《Machine Learning》
SE2324 《Mathematic Foundations for Computer Science》
CS0001W-01 《LLM Foundations and Practice》
Publications
Anti-Adversarial Learning: Desensitizing Prompts for Large Language Models
In Proceedings of the 40th Annual AAAI Conference on Artificial Intelligence (AAAI 2026). Singapore, Jan 20 - 27, 2026. (CCF-A)
From Code to Correctness: Closing the Last Mile of Code Generation with Hierarchical Debugging
In Proceedings of the 48th International Conference on Software Engineering (ICSE 2026). Rio De Janeiro, Brazil, April 12 - 18, 2026. (CCF-A)
LongCodeZip: Compress Long Context for Code Language Models
In Proceedings of the 40th International Conference on Automated Software Engineering (ASE 2025), Seoul, Korea, Nov. 16-20, 2025 (CCF-A)
Transplant Then Regenerate: A New Paradigm for Text Data Augmentation
In Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing (EMNLP 2025). Suzhou, China, Nov 5 - 9, 2025. (CCF-B)
LastingBench: Defend Benchmarks Against Knowledge Leakage
In Findings of the 2025 Conference on Empirical Methods in Natural Language Processing (EMNLP 2025 Findings). Suzhou, China, Nov 5 - 9, 2025. (CCF-B)
Between Lines of Code: Unraveling the Distinct Patterns of Machine and Human Programmers
In Proceedings of the 47th International Conference on Software Engineering (ICSE 2025). Ottawa, Ontario, Canada, April 27 - May 3, 2025. (CCF-A)
On the Effectiveness of Large Language Models in Domain-Specific Code Generation
ACM Transactions on Software Engineering and Methodology (TOSEM 2024) (CCF-A, ESI高被引论文)
How Effectively Do Code Language Models Understand Poor-Readability Code?
In Proceedings of the 39th ACM/IEEE International Conference on Automated Software Engineering (ASE 2024). Sacramento, California, United States, Oct 27 - Nov 1, 2024. (CCF-A)
VarGAN: Adversarial Learning of Variable Semantic Representations
IEEE Transactions on Software Engineering (TSE 2024) (CCF-A)
On the Evaluation of Neural Code Translation: Taxonomy and Benchmark
In Proceedings of the 38th International Conference on Automated Software Engineering (ASE 2023), Kirchberg, Luxembourg, Sept. 11-15, 2023 (CCF-A)
InfeRE: Step-by-Step Regex Generation via Chain of Inference
In Proceedings of the 38th International Conference on Automated Software Engineering (ASE 2023), Kirchberg, Luxembourg, Sept. 11-15, 2023 (CCF-A)
Self-Supervised Query Reformulation for Code Search
In Proceedings of the 31st ACM Joint European Software Engineering Conference and Symposium on the Foundations of Software Engineering (ESEC/FSE 2023), San Francisco, California, United States, Dec 3-9, 2023 (CCF-A)
Diet Code Is Healthy: Simplifying Programs for Pre-Trained Models of Code
In Proceedings of the 30th ACM Joint European Software Engineering Conference and Symposium on the Foundations of Software Engineering (ESEC/FSE 2022), Singapore, Nov 14-18, 2022 (CCF-A)
Cross-Domain Deep Code Search with Meta Learning
In Proceedings of the 44th International Conference on Software Engineering (ICSE 2022) (CCF-A)
Project Fund
上海自然科学基金面上项目，面向复杂场景的程序自动生成技术，2025.7-2028.6，主持
CCF-华为胡杨林基金，针对问题单解决的Multi-Agent能力提升,2025.1.1-2025.7.31，主持
华为，场景知识增强的Java代码自动生成技术，2024.9.1-2025.2.25，主持
国家重点研发计划，面向场景计算的低代码开发方法与环境，2023.12-2026.12，参与
中国航空无线电电子研究所，民机软件研制过程辅助系统，2022.12-2026.6，主持
上海交通大学-华为密码学与数字信任创新实验室课题，基于大模型的恶意代码样本生成，2023.5.1-2024.4.31，主持
国家自然科学基金，基于小样本学习的跨语言程序自动生成，2022.1.1-2024.12.31，主持
Address: No. 800 Dongchuan Road, Shanghai
Email: https://www.yzsfdd.com
Postal Code: 200240
Tel: +86-21-34205058 / 34204728
(Trial Version) Copyright © 2025 School of Computer Science, Shanghai Jiao Tong University. 沪交ICP备2010898
Chinese
