Elamur Research⌕
☰
home / papers / 2026 / AIRCoder: Adaptive Integration of Multi-dimensional Retrieval for Repository-level Code Completion

AIRCoder: Adaptive Integration of Multi-dimensional Retrieval for Repository-level Code Completion

Paper · 2026 · Chuanqi Shi, Miao Gao and Zhiqiang Gao

Summary

Repository-level code completion depends on pulling relevant context from very large codebases. Most current retrievers use only one signal, either textual similarity or dependency links, so their quality varies across benchmarks. Adaptive and hybrid alternatives add costly LLM reranking or retriever training, and none explicitly use the code's structural hierarchy. AIRCoder addresses this in two steps. First, files are split along AST boundaries, and each chunk is tagged with global metadata (file path, enclosing class and function signatures) and local metadata (signatures defined inside the chunk). Second, eight retrieval metrics are combined across three dimensions: sparse and dense textual similarity, dependency existence, and sparse and dense structural similarity. The combination is done either with training-free Reciprocal Rank Fusion (AIRCoder r) or with a small MLP that learns per-query weights (AIRCoder a). On CrossCodeEval and RepoEval, AIRCoder beats the strongest baseline in exact match and runs much faster than LLM-reranking methods. It also works across several programming languages without language-specific engineering.

Key points

  • Over the best baseline, AIRCoder gains 4.63% in exact match on average across CrossCodeEval and RepoEval. Qwen2.5-Coder-7B is among the generators evaluated.
  • Average latency is 0.366 s per query, versus 3.732 s for CodeRAG (about 10.2× faster) and 0.610 s for RepoCoder (about 1.7× faster). DraCo is quicker but limited to Python.
  • Dependencies are extracted with Qwen3-8B rather than static analysis. On CrossCodeEval-Python this yields 3,069 identifiers versus DraCo's 908, with 69.93% recall against DraCo and an 11.37% hallucination rate. This step accounts for 87.4% of AIRCoder's runtime.
  • On the Java, C# and TypeScript subsets of CrossCodeEval, AIRCoder a leads all baselines. On Java it beats RLCoder by 4.6% Code-EM.
  • Code is released at github.com/chuanky/AIRCoder.

Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.

 Institutions

🎓Southeast UniversityNanjing · China28 works · 354 citations🎓Southeast UniversityDhaka · Bangladesh8 works · 160 citations

 Fields, topics and concepts

📚 Shelves 5The field of science this work belongs to (OpenAlex fields).Computer Science · 16,667 worksArtificial Intelligence · 16,194 worksArts and Humanities · 16,039 worksLanguage and Linguistics · 15,891 worksInformation Systems · 1,437 works

🧭 Topics 1Research subjects and languages the work is about.Code & programming languages · 129 works

Cite

Cite as Chuanqi Shi, Miao Gao and Zhiqiang Gao. 2026. AIRCoder: Adaptive Integration of Multi-dimensional Retrieval for Repository-level Code Completion. Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers).

The paper itself, not this page. BibTeX from the library's record of the work. DOI 10.18653/v1/2026.acl-long.1166

@inproceedings{shi2026aircoder,
  title = {{AIRCoder: Adaptive Integration of Multi-dimensional Retrieval for Repository-level Code Completion}},
  author = {Chuanqi Shi, Miao Gao and Zhiqiang Gao},
  booktitle = {Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)},
  year = {2026},
  doi = {10.18653/v1/2026.acl-long.1166},
  url = {https://doi.org/10.18653/v1/2026.acl-long.1166}
}

References

34 references, 12 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Jacob Austin, Augustus Odena, Maxwell Nye, Maarten Bosma, Henryk Michalewski, David Dohan, Ellen Jiang, Carrie Cai, Michael Terry, Quoc Le, and 1 others. 2021. Program synthesis with large language models. arXiv preprint arXiv:2108.07732.
  2. Mark Chen, Jerry Tworek, Heewoo Jun, Qiming Yuan, Henrique Pondé de Oliveira Pinto, Jared Kaplan, Harri Edwards, Yuri Burda, Nicholas Joseph, Greg Brockman, Alex Ray, Raul Puri, Gretchen Krueger, Michael Petrov, Heidy Khlaaf, Girish Sastry, Pamela Mishkin, Brooke Chan, Scott Gray, and 39 others. 202… — Evaluating Large Language Models Trained on Code
  3. Wei Cheng, Yuhan Wu, and Wei Hu. 2024. Dataflow-guided retrieval augmentation for repository-level code completion. In Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), ACL 2024, Bangkok, Thailand, August 11-16, 2024, pages 7957– 7977. A… find
  4. Gordon V. Cormack, Charles L A Clarke, and Stefan Buettcher. 2009. Reciprocal rank fusion outperforms condorcet and individual rank learning methods. In Proceedings of the 32nd International ACM SIGIR Conference on Research and Development in Information Retrieval, SIGIR ’09, page 758–759, New York,…
  5. Yangruibo Ding, Zijian Wang, Wasi Uddin Ahmad, Hantian Ding, Ming Tan, Nihal Jain, Murali Krishna Ramanathan, Ramesh Nallapati, Parminder Bhatia, Dan Roth, and Bing Xiang. 2023. Crosscodeeval: A diverse and multilingual benchmark for cross-file code completion. In Advances in Neural Information Proc… find
  6. Yangruibo Ding, Zijian Wang, Wasi Uddin Ahmad, Murali Krishna Ramanathan, Ramesh Nallapati, Parminder Bhatia, Dan Roth, and Bing Xiang. 2024. Cocomic: Code completion by jointly modeling in-file and cross-file context. In Proceedings of the 2024 Joint International Conference on Computational Lingui…
  7. Fabio James Fehr, Prabhu Teja S, Luca Franceschi, and Giovanni Zappella. 2025. CoRet: Improved retriever for code editing. In Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers), pages 775–789, Vienna, Austria. Association for Computationa… find
  8. Daya Guo, Shuai Lu, Nan Duan, Yanlin Wang, Ming Zhou, and Jian Yin. 2022. Unixcoder: Unified cross-modal pre-training for code representation. In Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), ACL 2022, Dublin, Ireland, May 22-27, 202…
  9. Daya Guo, Qihao Zhu, Dejian Yang, Zhenda Xie, Kai Dong, Wentao Zhang, Guanting Chen, Xiao Bi, Y. Wu, Y. K. Li, Fuli Luo, Yingfei Xiong, and Wenfeng Liang. 2024. Deepseek-coder: When the large language model meets programming - the rise of code intelligence. CoRR, abs/2401.14196. find
  10. Paul Jaccard. 1912. The distribution of the flora in the alpine zone. The New Phytologist, 11(2):37–50. find
  11. Kin Kwan Leung, Mouloud Belbahri, Yi Sui, Alex Labach, Xueying Zhang, Stephen Rose, and Jesse C. Cresswell. 2025. Classifying and addressing the diversity of errors in retrieval-augmented generation systems. Preprint, arXiv:2510.13975. find
  12. VI Levenshtein. 1966. Binary coors capable or ‘correcting deletions, insertions, and reversals. In Soviet physics-doklady, volume 10. find
  13. Patrick Lewis, Ethan Perez, Aleksandra Piktus, Fabio Petroni, Vladimir Karpukhin, Naman Goyal, Heinrich Küttler, Mike Lewis, Wen-tau Yih, Tim Rocktäschel, and 1 others. 2020. Retrieval-augmented generation for knowledge-intensive nlp tasks. Advances in neural information processing systems, 33:9459–…
  14. Ming Liang, Xiaoheng Xie, Gehao Zhang, Xunjin Zheng, Peng Di, Wei Jiang, Hongwei Chen, Chengpeng Wang, and Gang Fan. 2024. REPOFUSE: repository-level code completion with fused dual context. CoRR, abs/2402.14323. — [arXiv:2402.14323](https://arxiv.org/abs/2402.14323) find
  15. Wei Liu, Ailun Yu, Daoguang Zan, Bo Shen, Wei Zhang, Haiyan Zhao, Zhi Jin, and Qianxiang Wang. 2024. Graphcoder: Enhancing repository-level code completion via coarse-to-fine retrieval based on code context graph. In Proceedings of the 39th IEEE/ACM International Conference on Automated Software Eng… find
  16. Anton Lozhkov, Raymond Li, Loubna Ben Allal, Federico Cassano, Joel Lamy-Poirier, Nouamane Tazi, Ao Tang, Dmytro Pykhtar, Jiawei Liu, Yuxiang Wei, Tianyang Liu, Max Tian, Denis Kocetkov, Arthur Zucker, Younes Belkada, Zijian Wang, Qian Liu, Dmitry Abulkhanov, Indraneil Paul, and 38 others. 2024. Sta… find
  17. Hou Min and Zhang Li Ping. 2019. Survey on software clone detection research. In Proceedings of the 2019 3rd International Conference on Management Engineering, Software Engineering and Service Sciences, ICMSS 2019, page 9–16, New York, NY, USA. Association for Computing Machinery. find
  18. Anh Nguyen-Duc, Beatriz Cabrero-Daniel, Adam Przy-bylek, Chetan Arora, Dron Khanna, Tomas Herda, Usman Rafiq, Jorge Melegati, Eduardo Guerra, Kai-Kristian Kemell, and 1 others. 2025. Generative artificial intelligence for software engineering—a research agenda. Software: Practice and Experience, 55(… find
  19. Erik Nijkamp, Bo Pang, Hiroaki Hayashi, Lifu Tu, Huan Wang, Yingbo Zhou, Silvio Savarese, and Caiming Xiong. 2023. Codegen: An open large language model for code with multi-turn program synthesis. In The Eleventh International Conference on Learning Representations, ICLR 2023, Kigali, Rwanda, May 1-… find
  20. Md. Rizwan Parvez, Wasi Uddin Ahmad, Saikat Chakraborty, Baishakhi Ray, and Kai-Wei Chang. 2021. Retrieval augmented code generation and summarization. In Findings of the Association for Computational Linguistics: EMNLP 2021, Virtual Event / Punta Cana, Dominican Republic, 16-20 November, 2021, page…
  21. YIBO Peng, Zora Zhiruo Wang, and Daniel Fried. 2025. Can long-context language models solve repository-level code generation? In LTI Student Research Symposium 2025. find
  22. Ori Ram, Yoav Levine, Itay Dalmedigos, Dor Muhlgay, Amnon Shashua, Kevin Leyton-Brown, and Yoav Shoham. 2023. In-context retrieval-augmented language models. Trans. Assoc. Comput. Linguistics, 11:1316–1331.
  23. Baptiste Rozière, Jonas Gehring, Fabian Gloeckle, Sten Sootla, Itai Gat, Xiaoqing Ellen Tan, Yossi Adi, Jingyu Liu, Tal Remez, Jérémy Rapin, Artyom Kozhevnikov, Ivan Evtimov, Joanna Bitton, Manish Bhatt, Cristian Canton-Ferrer, Aaron Grattafiori, Wenhan Xiong, Alexandre Défossez, Jade Copet, and 6 o… — Code Llama: Open Foundation Models for Code
  24. Freda Shi, Xinyun Chen, Kanishka Misra, Nathan Scales, David Dohan, Ed H. Chi, Nathanael Schärli, and Denny Zhou. 2023. Large language models can be easily distracted by irrelevant context. In International Conference on Machine Learning, ICML 2023, 23-29 July 2023, Honolulu, Hawaii, USA, volume 202…
  25. Weijia Shi, Sewon Min, Michihiro Yasunaga, Minjoon Seo, Richard James, Mike Lewis, Luke Zettlemoyer, and Wen-tau Yih. 2024. REPLUG: retrieval-augmented black-box language models. In Proceedings of the 2024 Conference of the North American Chapter of the Association for Computational Linguistics: Hum…
  26. Yuling Shi, Yichun Qian, Hongyu Zhang, Beijun Shen, and Xiaodong Gu. 2025. Longcodezip: Compress long context for code language models. CoRR, abs/2510.00446. — [arXiv:2510.00446](https://arxiv.org/abs/2510.00446) find
  27. Disha Shrivastava, Denis Kocetkov, Harm de Vries, Dzmitry Bahdanau, and Torsten Scholak. 2023a. Repofusion: Training code models to understand your repository. CoRR, abs/2306.10998. find
  28. Disha Shrivastava, Hugo Larochelle, and Daniel Tarlow. 2023b. Repository-level prompt generation for large language models of code. In International Conference on Machine Learning, ICML 2023, 23-29 July 2023, Honolulu, Hawaii, USA, volume 202 of Proceedings of Machine Learning Research, pages 31693–… find
  29. Hanzhuo Tan, Qi Luo, Ling Jiang, Zizheng Zhan, Jing Li, Haotian Zhang, and Yuqun Zhang. 2024. Prompt-based code completion via multi-retrieval augmented generation. CoRR, abs/2405.07530. — [arXiv:2405.07530](https://arxiv.org/abs/2405.07530) find
  30. Dmitry Ustalov, Egor Bogomolov, Alexander Bez-zubov, Yaroslav Golubev, Evgeniy Glukhov, Georgii Levtsov, and Vladimir Kovalenko. 2025. Challenge on optimization of context collection for code completion. CoRR, abs/2510.04349. — [arXiv:2510.04349](https://arxiv.org/abs/2510.04349) find
  31. Yanlin Wang, Yanli Wang, Daya Guo, Jiachi Chen, Ruikai Zhang, Yuchi Ma, and Zibin Zheng. 2025. Rlcoder: Reinforcement learning for repository-level code completion. In 47th IEEE/ACM International Conference on Software Engineering, ICSE 2025, Ottawa, ON, Canada, April 26 - May 6, 2025, pages 1140–11… find
  32. Di Wu, Wasi Uddin Ahmad, Dejiao Zhang, Murali Krishna Ramanathan, and Xiaofei Ma. 2024. Repoformer: Selective retrieval for repository-level code completion. In Forty-first International Conference on Machine Learning, ICML 2024, Vienna, Austria, July 21-27, 2024. OpenReview.net. find
  33. Ori Yoran, Tomer Wolfson, Ori Ram, and Jonathan Berant. 2024. Making retrieval-augmented language models robust to irrelevant context. In The Twelfth International Conference on Learning Representations, ICLR 2024, Vienna, Austria, May 7-11, 2024. OpenReview.net.
  34. Fengji Zhang, Bei Chen, Yue Zhang, Jacky Keung, Jin Liu, Daoguang Zan, Yi Mao, Jian-Guang Lou, and Weizhu Chen. 2023. Repocoder: Repository-level code completion through iterative retrieval and generation. In Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing, EMN… find

Updated

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/shi-2026-aircoder-adaptive-integration-multi-dimensional/