Publications

(#: Corresponding Author, *: Equal Contribution)

Preprint

  • [Preprint] Escalating LLM-based Code Translation Benchmarking into the Class-level Era
    Pengyu Xue, Linhao Wu, Chengyi Wang, Xiang Li, Zhen Yang, Ruikai Jin, Yuxiang Zhang, Jia Li, Yifei Pei, Zhaoyan Shen, Xiran Lyu
    Paper

  • [Preprint] Generating Equivalent Representations of Code By A Self-Reflection Approach
    Jia Li, Ge Li, Lecheng Wang, Hao Zhu, Zhi Jin
    Paper

  • [Preprint] Showing LLM-Generated Code Selectively Based on Confidence of LLMs
    Jia Li, Yuqi Zhu, Yongmin Li, Ge Li, Zhi Jin
    Paper

  • [Preprint] FAN: Fourier Analysis Networks
    Yihong Dong*, Ge Li*, Yongding Tao, Xue Jiang, Kechi Zhang, Jia Li, Jing Su, Jun Zhang, Jingjing Xu
    Paper

  • [Preprint] Code Structure-Aware through Line-level Semantic Learning for Code Vulnerability Detection
    Ziliang Wang, Ge Li, Jia Li, Yihong Dong, Yingfei Xiong, Zhi Jin
    Paper

  • [Preprint] ChatCoder: Chat-based Refine Requirement Improves LLMs' Code Generation
    Zejun Wang, Jia Li, Ge Li, Zhi Jin
    Paper

  • [Preprint] Large Language Model-Aware In-Context Learning for Code Generation
    Jia Li, Ge Li, Chongyang Tao, Jia Li, Huangzhao Zhang, Fang Liu, Zhi Jin
    Paper

  • [Preprint] ToolCoder: Teach Code Generation Models to use API search tools
    Kechi Zhang, Huangzhao Zhang, Ge Li, Jia Li, Zhuo Li, Zhi Jin
    Paper

2025
  • [ICSE 2025] aiXcoder-7B: A Lightweight and Effective Large Language Model for Code Completion CCF-A
    Siyuan Jiang*, Jia Li*, He Zong, Huanyu Liu, Hao Zhu, Shukai Hu, Erlu Li, Jiazheng Ding, Yu Han, Wei Ning, Ge Li
    In: The 47th International Conference on Software Engineering (ICSE), April 30-May 2, 2025, Ottawa, Canada.
    (Siyuan Jiang and Jia Li contribute equally and are co-first authors.)
    Paper Model

2024
  • [NeurIPS 2024] EvoCodeBench: An Evolving Code Generation Benchmark with Domain-Specific Evaluations CCF-A
    Jia Li, Ge Li, Xuanming Zhang, Yihong Dong, Zhi Jin, Binhua Li, Fei Huang, Yongbing Li
    In: The Thirty-eight Conference on Neural Information Processing Systems (NeurIPS), December 10-15, 2024, Vancouver, Canada.
    Paper Data

  • [TOSEM] Structured Chain-of-Thought Prompting for Code GenerationCCF-A
    Jia Li, Yongmin Li, Ge Li, Zhi Jin
    In: ACM Transactions on Software Engineering and Methodology (TOSEM), 2024.
    Paper

  • [TOSEM] AceCoder: An Effective Prompting Technique Specialized in Code Generation CCF-A
    Jia Li, Yunfei Zhao, Yongmin Li, Ge Li#, Zhi Jin
    In: ACM Transactions on Software Engineering and Methodology (TOSEM), 2024.
    Paper

  • [ACL 2024] DevEval: A Manually-Annotated Code Generation Benchmark Aligned with Real-World Code Repositories CCF-A
    Jia Li, Ge Li#, Yunfei Zhao, Yongmin Li, Huanyu Liu, Hao Zhu, Lecheng Wang, Kaibo Liu, Zheng Fang, Lansheng Wang, Jiazheng, Ding, Xuanming Zhang, Yuqi Zhu, Yihong Dong, Zhi Jin, Binhua Li, Fei Huang, Yongbing Li#
    In: The 62st Annual Meeting of the Association for Computational Linguistics (ACL), Augest 11-16, 2024, Bangkok, Thailand.
    Paper Data

  • [FSE 2024] Exploring and Unleashing the Power of Large Language Models in Automated Code Translation CCF-A
    Zhen Yang, Fang Liu#, Zhongxing Yu, Jacky Wai Keung, Jia Li, Shuo Liu, Yifan Hong, Xiaoxue Ma, Ge Li, Zhi Jin
    In: The ACM International Conference on the Foundations of Software Engineering (FSE), July 17-19, 2024.
    Paper Souce Code

  • [SCIS] Deep Learning for Code Generation: A Survey CCF-A
    Huangzhao Zhang, Kechi Zhang, Zhuo Li, Jia Li, Jia Li, Yongmin Li, Yunfei Zhao, Yuqi Zhu, Fang Liu, Ge Li, Zhi Jin#
    In: Science China Information Sciences (SCIS), 2024
    Paper

  • [AAAI 2024] Hot or Cold? Adaptive Temperature Sampling for Code Generation with Large Language Models CCF-A
    Yuqi Zhu, Jia Li, Ge Li, Yunfei Zhao, Jia Li, Zhi Jin, Hong Mei
    In: The 38th Annual AAAI Conference on Artificial Intelligence (AAAI), February 20-27, 2024, Vancouver, Canada.
    Acceptance Rate: 23.75% (2342/9862).
    Paper Souce Code

2023
  • [TOSEM] Poison Attack and Poison Detection on Deep Source Code Processing Models CCF-A
    Jia Li, Zhuo Li, Huangzhao Zhang, Ge Li, Zhi Jin, Xing Hu, Xin Xia
    In: ACM Transactions on Software Engineering and Methodology (TOSEM), 2023.
    Paper

  • [ASE 2023] ZC3 Zero-Shot Cross-Language Code Clone Detection CCF-A
    Jia Li, Chongyang Tao, Zhi Jin#, Fang Liu, Jia Li, Ge Li#
    In: The 38th IEEE/ACM International Conference on Automated Software Engineering (ASE), September 11-15, 2023, Kirchberg, Luxembourg.
    Acceptance Rate: 21% (134/629).
    Paper Souce Code

  • [Internetware 2023] MCodeSearcher Multi-View Contrastive Learning for Code Search CCF-C
    Jia Li, Fang Liu, Jia Li, Yunfei Zhao, Ge Li#, Zhi Jin#
    In: The 14th Asia-Pacific Symposium on Internetware, August 4-6, 2023, Hangzhou, China.
    Acceptance Rate: 40% (30/75).
    Paper

  • [ACL 2023] Self-Edit: Fault-Aware Code Editor for Code GenerationCCF-A
    Kechi Zhang, Zhuo Li, Jia Li, Ge Li#, Zhi Jin#
    In: The 61st Annual Meeting of the Association for Computational Linguistics (ACL), July 9-14, 2023, Toronto, Canada.
    Acceptance Rate: 22.7%.
    Paper Source Code

  • [TOSEM] CodeEditor: Learning to Edit Source Code with Pre-trained Models CCF-A
    Jia Li, Ge Li#, Zhuo Li, Zhi Jin#, Xing Hu, Kechi Zhang, Zhiyi Fu
    In: ACM Transactions on Software Engineering and Methodology (TOSEM), 2023.
    (Selected for ASE 2023 journal-first presentation)
    Paper Source Code

  • [ICSE 2023] SkCoder: A Sketch-based Approach for Automatic Code Generation CCF-A
    Jia Li, Yongmin Li, Ge Li#, Zhi Jin#
    In: The 45th IEEE/ACM International Conference on Software Engineering (ICSE), May 17-19, 2023, Melbourne, Australia.
    Acceptance Rate: 26% (209/796).
    Paper Source Code

  • [NeurIPS 2022] Fine-Tuning Pre-Trained Language Models Effectively by Optimizing Subnetworks Adaptively CCF-A
    Haojie Zhang, Ge Li#, Jia Li, Zhongjin Zhang, Yuqi Zhu, Zhi Jin#
    In: The 36th Conference on Neural Information Processing Systems (NeurIPS), November 28-December 9, 2023, virtual.
    Acceptance Rate: 25.6%.
    Paper Source Code

2021
  • [ASE 2021] EditSum: A Retrieve-and-Edit Framework for Source Code Summarization CCF-A
    Jia Li, Yongmin Li, Ge Li#, Xing Hu, Xin Xia, Zhi Jin#
    In: The 36th IEEE/ACM International Conference on Automated Software Engineering (ASE), November 14-20, 2021, Australia.
    Acceptance Rate: 18.6% (82/440).
    Paper Source Code