Elamur Research⌕
☰
home / papers / 2026 / Contract-Coding: Towards Repo-Level Generation via Structured Symbolic Paradigm

Contract-Coding: Towards Repo-Level Generation via Structured Symbolic Paradigm

Paper · 2026 · Yilin, Lujin Zhao and Yijie Shi

Summary

The paper addresses a weakness of LLM agents asked to build whole software repositories from loose, informal requests ("Vibe Coding"). In sequential multi-agent pipelines such as MetaGPT and ChatDev, early design mistakes carry forward into later steps. The accumulating code also fills the context window, so the overall architecture eventually breaks down.

The authors propose Contract-Coding. An agent first converts the vague intent into a structured Language Contract, which acts as a Single Source of Truth (SSOT). The contract has a natural-language layer and an executable kernel that maps modules to files, type signatures and state. It is edited only through ADD and UPDATE operations, and any edit that breaks acyclicity or type safety is rejected. Because each file is generated from the contract alone, files can be written in parallel, and a contract-guided auditing step can repair the contract itself when it is wrong.

On the Greenfield-5 benchmark, the method reaches 47% functional success with near-perfect structural integrity under a 16k-token budget. This suggests that explicit architectural decoupling may scale better than simply enlarging the context window.

Key points

  • The framework runs in three phases: Symbolic Projection, Topological Orchestration through a Contract-Driven Hierarchical Graph (HEG), and Active Contract Auditing.
  • Greenfield-5 consists of curated game-engine tasks. Its hardest task, Roguelike, needs 15–25 interdependent files.
  • Contract-Coding achieved 47% functional success and 100% Structural Integrity with a 16k context limit. The commercial Gemini Studio reached 63% but relied on more than 200k tokens.
  • The baselines (MetaGPT, ChatDev, FLOW, OpenHands, Lingma) failed in different ways, including grounding failures, context collapse and recursive loops. The authors also report up to 4.6× token compression, with contract size growing sub-linearly relative to repository size.
  • The stated limitations are an evaluation of only five tasks, plus extra latency and token cost from the auditing loop. Code is available at github.com/imliinyi/Contract-Coding.

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

🎓Beijing University of Posts and TelecommunicationsBeijing · China78 works · 33,365 citations🔬State Key Laboratory of Networking and Switching TechnologyBeijing · China9 works · 273 citations

 Fields, topics and concepts

📚 Shelves 3The field of science this work belongs to (OpenAlex fields).Computer Science · 18,614 worksArtificial Intelligence · 17,965 worksInformation Systems · 1,730 works

Cite

Cite as Yilin, Lujin Zhao and Yijie Shi. 2026. Contract-Coding: Towards Repo-Level Generation via Structured Symbolic Paradigm. Findings of the Association for Computational Linguistics: ACL 2026.

The paper itself, not this page. BibTeX from the library's record of the work. DOI 10.18653/v1/2026.findings-acl.400

@inproceedings{yilin2026contract,
  title = {{Contract-Coding: Towards Repo-Level Generation via Structured Symbolic Paradigm}},
  author = {Yilin, Lujin Zhao and Yijie Shi},
  booktitle = {Findings of the Association for Computational Linguistics: ACL 2026},
  year = {2026},
  doi = {10.18653/v1/2026.findings-acl.400},
  url = {https://doi.org/10.18653/v1/2026.findings-acl.400}
}

References

41 references, 26 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Mark Chen, Jerry Tworek, Heewoo Jun, Qiming Yuan, Henrique Ponde 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
  2. Qian Chen, Liu Wei, Liu Hongzhang, Chen Nuo, Dang Yufan, Li Jiahao, Yang Cheng, Chen Weize, Su Yusheng, Cong Xin, Xu Juyuan, Li Dahai, Liu Zhiyuan, and Sun Maosong. 2024. ChatDev: Communicative agents for software development. In Proceedings of the 62nd Annual Meeting of the Association for Computat…
  3. Weize Chen, Yusheng Su, Jingwei Zuo, Cheng Yang, Chenfei Yuan, Chi-Min Chan, Heyang Yu, Yaxi Lu, Yi-Hsin Hung, Chen Qian, Yujia Qin, Xin Cong, Ruobing Xie, Zhiyuan Liu, Maosong Sun, and Jie Zhou. 2023. Agentverse: Facilitating multi-agent collaboration and exploring emergent behaviors. Preprint, arX…
  4. Marone, Christopher Akiki, Jia Li, Jenny Chim, Qian Liu, Evgenii Zheltonozhskii, Terry Yue Zhuo, Thomas Wang, Olivier Dehaene, Mishig Davaadorj, Joel Lamy-Poirier, João Monteiro, Oleh Shliazhko, and 48 others. 2023b. Starcoder: may the source be with you! Preprint, arXiv:2305.06161. find
  5. Yujia Li, David Choi, Junyoung Chung, Nate Kushman, Julian Schrittwieser, Rémi Leblond, Tom Eccles, James Keeling, Felix Gimeno, Agustin Dal Lago, Thomas Hubert, Peter Choy, Cyprien de Masson d’Autume, Igor Babuschkin, Xinyun Chen, Po-Sen Huang, Johannes Welbl, Sven Gowal, Alexey Cherepanov, and 7 o… — Competition-Level Code Generation with AlphaCode
  6. Zelong Li, Shuyuan Xu, Kai Mei, Wenyue Hua, Balaji Rama, Om Raheja, Hao Wang, He Zhu, and Yongfeng Zhang. 2024. Autoflow: Automated workflow generation for large language model agents. CoRR, abs/2407.12821. — [arXiv:2407.12821](https://arxiv.org/abs/2407.12821) find
  7. Yi Lin, Lujin Zhao, and Yijie Shi. 2025. (p)rior(d)yna(f)low: A priori dynamic workflow construction via multi-agent collaboration. Preprint, arXiv:2509.14547. — [arXiv:2509.14547](https://arxiv.org/abs/2509.14547) find
  8. Hanxing Ding, Shuchang Tao, Liang Pang, Zihao Wei, Jinyang Gao, Bolin Ding, Huawei Shen, and Xueqi Cheng. 2025. Toolcoder: A systematic code-empowered tool learning framework for large language models. Preprint, arXiv:2502.11404. find
  9. Nelson F. Liu, Kevin Lin, John Hewitt, Ashwin Paranjape, Michele Bevilacqua, Fabio Petroni, and Percy Liang. 2024a. Lost in the middle: How language models use long contexts. Transactions of the Association for Computational Linguistics, 12:157–173.
  10. 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. Preprint, arXiv:2401.14196. find
  11. Zijun Liu, Yanzhe Zhang, Peng Li, Yang Liu, and Diyi Yang. 2024b. A dynamic LLM-powered agent network for task-oriented agent collaboration. In First Conference on Language Modeling.
  12. Sirui Hong, Mingchen Zhuge, Jonathan Chen, Xiawu Zheng, Yuheng Cheng, Jinlin Wang, Ceyao Zhang, Zili Wang, Steven Ka Shing Yau, Zijuan Lin, Liyang Zhou, Chenyu Ran, Lingfeng Xiao, Chenglin Wu, and Jürgen Schmidhuber. 2024. MetaGPT: Meta programming for a multi-agent collaborative framework. In The T…
  13. Carlos E Jimenez, John Yang, Alexander Wettig, Shunyu Yao, Kexin Pei, Ofir Press, and Karthik R Narasimhan. 2024. SWE-bench: Can language models resolve real-world github issues? In The Twelfth International Conference on Learning Representations.
  14. Guohao Li, Hasan Abed Al Kader Hammoud, Hani Itani, Dmitrii Khizbullin, and Bernard Ghanem. 2023a. Camel: Communicative agents for "mind" exploration of large language model society. Preprint, arXiv:2303.17760.
  15. Raymond Li, Loubna Ben Allal, Yangtian Zi, Niklas Muennighoff, Denis Kocetkov, Chenghao Mou, Marc George Ma, Anurag Koul, Qi Chen, Yawen Wu, Sachit Kuhar, Yu Yu, Aritra Sengupta, Varun Kumar, and Murali Krishna Ramanathan. 2025. Specagent: A speculative retrieval and forecasting agent for code compl…
  16. Zhenyu Mao, Jacky Keung, Fengji Zhang, Shuo Liu, Yifei Wang, and Jialong Li. 2025. Towards engineering multi-agent llms: A protocol-driven approach. Preprint, arXiv:2510.12120. — [arXiv:2510.12120](https://arxiv.org/abs/2510.12120) find
  17. B. Meyer. 1992. Applying ’design by contract’. Computer, 25(10):40–51. find
  18. Boye Niu, Yiliao Song, Kai Lian, Yifan Shen, Yu Yao, Kun Zhang, and Tongliang Liu. 2025. Flow: Modularized agentic workflow automation. In The Thirteenth International Conference on Learning Representations. find
  19. Shinn Noah, Cassano Federico, Gopinath Ashwin, Narasimhan Karthik, and Yao Shunyu. 2023. Reflexion: language agents with verbal reinforcement learning. In Advances in Neural Information Processing Systems, volume 36, pages 8634–8652. Curran Associates, Inc.
  20. OpenAI, Josh Achiam, Steven Adler, Sandhini Agarwal, Lama Ahmad, Ilge Akkaya, Florencia Leoni Aleman, Diogo Almeida, Janko Altenschmidt, Sam Altman, Shyamal Anadkat, Red Avila, Igor Babuschkin, Suchir Balaji, Valerie Balcom, Paul Baltescu, Haiming Bao, Mohammad Bavarian, Jeff Belgum, and 262 others.… — GPT-4 Technical Report
  21. Charles Packer, Sarah Wooders, Kevin Lin, Vivian Fang, Shishir G. Patil, Ion Stoica, and Joseph E. Gonzalez. 2024. Memgpt: Towards llms as operating systems. Preprint, arXiv:2310.08560. find
  22. D. L. Parnas. 1972. On the criteria to be used in decomposing systems into modules. Commun. ACM, 15(12):1053–1058. find
  23. Shishir G. Patil, Tianjun Zhang, Xin Wang, and Joseph E. Gonzalez. 2023. Gorilla: Large language model connected with massive apis. Preprint, arXiv:2305.15334.
  24. Baptiste Rozière, Jonas Gehring, Fabian Gloeckle, Sten Sootla, Itai Gat, Xiaoqing Ellen Tan, Yossi Adi, Jingyu Liu, Romain Sauvestre, Tal Remez, Jérémy Rapin, Artyom Kozhevnikov, Ivan Evtimov, Joanna Bitton, Manish Bhatt, Cristian Canton Ferrer, Aaron Grattafiori, Wenhan Xiong, Alexandre Défossez, a… — Code Llama: Open Foundation Models for Code
  25. Timo Schick, Jane Dwivedi-Yu, Roberto Dessì, Roberta Raileanu, Maria Lomeli, Luke Zettlemoyer, Nicola Cancedda, and Thomas Scialom. 2023. Toolformer: Language models can teach themselves to use tools. Preprint, arXiv:2302.04761.
  26. Yongliang Shen, Kaitao Song, Xu Tan, Dongsheng Li, Weiming Lu, and Yueting Zhuang. 2023. Hugginggpt: Solving ai tasks with chatgpt and its friends in hugging face. Preprint, arXiv:2303.17580.
  27. Amit Sheth, Kaushik Roy, and Manas Gaur. 2023. Neurosymbolic ai – why, what, and how. Preprint, arXiv:2305.00813. — [arXiv:2305.00813](https://arxiv.org/abs/2305.00813) find
  28. Yashar Talebirad and Amirhossein Nadiri. 2023. Multi-agent collaboration: Harnessing the power of intelligent llm agents. Preprint, arXiv:2306.03314.
  29. Lei Wang, Chen Ma, Xueyang Feng, Zeyu Zhang, Hao Yang, Jingsen Zhang, Zhiyuan Chen, Jiakai Tang, Xu Chen, Yankai Lin, Wayne Xin Zhao, Zhewei Wei, and Jirong Wen. 2024a. A survey on large language model based autonomous agents. Frontiers of Computer Science, 18(6).
  30. Xingyao Wang, Boxuan Li, Yufan Song, Frank F. Xu, Xiangru Tang, Mingchen Zhuge, Jiayi Pan, Yueqi Song, Bowen Li, Jaskirat Singh, Hoang H. Tran, Fuqiang Li, Ren Ma, Mingzhang Zheng, Bill Qian, Yanjun Shao, Niklas Muennighoff, Yizhe Zhang, Binyuan Hui, and 5 others. 2025. Openhands: An open platform f…
  31. Zhenhailong Wang, Shaoguang Mao, Wenshan Wu, Tao Ge, Furu Wei, and Heng Ji. 2024b. Unleashing the emergent cognitive synergy in large language models: A task-solving agent through multi-persona self-collaboration. Preprint, arXiv:2307.05300.
  32. Jason Wei, Xuezhi Wang, Dale Schuurmans, Maarten Bosma, brian ichter, Fei Xia, Ed H. Chi, Quoc V Le, and Denny Zhou. 2022. Chain of thought prompting elicits reasoning in large language models. In Advances in Neural Information Processing Systems.
  33. Jiaxin Wen, Jian Guan, Hongning Wang, Wei Wu, and Minlie Huang. 2025. Codeplan: Unlocking reasoning potential in large language models by scaling code-form planning. In The Thirteenth International Conference on Learning Representations. find
  34. Zhiheng Xi, Wenxiang Chen, Xin Guo, Wei He, Yiwen Ding, Boyang Hong, Ming Zhang, Junzhe Wang, Senjie Jin, Enyu Zhou, Rui Zheng, Xiaoran Fan, Xiao Wang, Limao Xiong, Yuhao Zhou, Weiran Wang, Changhao Jiang, Yicheng Zou, Xiangyang Liu, and 10 others. 2023. The rise and potential of large language mode…
  35. Yiheng Xu, Hongjin SU, Chen Xing, Boyu Mi, Qian Liu, Weijia Shi, Binyuan Hui, Fan Zhou, Yitao Liu, Tianbao Xie, Zhoujun Cheng, Siheng Zhao, Lingpeng Kong, Bailin Wang, Caiming Xiong, and Tao Yu. 2024. Lemur: Harmonizing natural language and code for language agents. In The Twelfth International Conf… find
  36. Hui Yang, Sifu Yue, and Yunzhong He. 2023. Autogpt for online decision making: Benchmarks and additional opinions. Preprint, arXiv:2306.02224.
  37. John Yang, Carlos E. Jimenez, Alexander Wettig, Kilian Lieret, Shunyu Yao, Karthik Narasimhan, and Ofir Press. 2024. Swe-agent: Agent-computer interfaces enable automated software engineering. Preprint, arXiv:2405.15793.
  38. Eric Zelikman, Qian Huang, Gabriel Poesia, Noah Goodman, and Nick Haber. 2023. Parsel: Algorithmic reasoning with language models by composing decompositions. In Advances in Neural Information Processing Systems, volume 36, pages 31466–31523. Curran Associates, Inc.
  39. Guibin Zhang, Yanwei Yue, Zhixun Li, Sukwon Yun, Guancheng Wan, Kun Wang, Dawei Cheng, Jeffrey Xu Yu, and Tianlong Chen. 2025. Cut the crap: An economical communication pipeline for LLM-based multi-agent systems. In The Thirteenth International Conference on Learning Representations. find
  40. Kechi Zhang, Jia Li, Ge Li, Xianjie Shi, and Zhi Jin. 2024. Codeagent: Enhancing code generation with tool-integrated agent systems for real-world repo-level coding challenges. ACL.
  41. Qianhui Zhao, Li Zhang, Fang Liu, Junhang Cheng, Chengru Wu, Junchen Ai, Qiaoyuanhe Meng, Lichen Zhang, Xiaoli Lian, Shubin Song, and Yuanping Guo. 2025. Towards realistic project-level code generation via multi-agent collaboration and semantic architecture modeling. Preprint, arXiv:2511.03404. — [arXiv:2511.03404](https://arxiv.org/abs/2511.03404) find

Updated

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/yilin-2026-contract-coding-repo-level-generation/