Repository-Level Code Generation

Follow this topic Ask about this topic

Repository-level code generation studies how AI systems can implement features, fix issues, or build projects while respecting the code, interfaces, and conventions already present. Common approaches retrieve relevant files with semantic search or dependency graphs, organize requirements into specifications or contracts, and coordinate specialized agents; testing and static analysis provide feedback for refinement. Benchmarks evaluate both repository repair and projects built from scratch, often using hidden tests and real execution environments. A central challenge is that passing tests does not guarantee sound architecture or maintainability, while agents must also find useful, current context without spending excessive time or tokens exploring a codebase.

Latest papers

43 papers7 papers in the last four weeks, up 40% on the four weeks before. 0.0% of all new papers.

Skill Dreaming: Evolving Repository Skills through Pull Request Streams
Oct 26, 2026Xinyuan Song, Hasibul Haque, Jay Somasundaram +3Automated Software Engineering LLM Agent Skill Learning
Causal Dynamics Lab, USA
When LLM-Generated Refactorings Ripple: An Empirical Study of Repository Adoption Readiness
Oct 17, 2026Muhammad Waseem, Aakash Ahmad, Pekka AbrahamssonAutomated Code Refactoring LLM Reliability
Tampere University, Finland. · the School of Computing and Engineering, University of Derby, UK.
OPRD: On-Policy Reminder Distillation for Long-Horizon Coding Agents
Oct 15, 2026Yuechen Luo, Zhiyi Lai, Miao Ziqi +9AI Coding Agents LLM Agent Training
Tsinghua University · Independent Researcher · Shanghai Artificial Intelligence Laboratory +1
Chronos Enables Code Agents to Reason over Software Evolution
Oct 8, 2026Xin Yin, Yiang Zhang, Zhiyuan Peng +3LLM Agent Memory Repository-Level Code Generation
Zhejiang University, China · National Industrial Information Security Development Research Center, China
Large-scale Repository Engineering via Agent-Native Reusable Code Primitives
Oct 6, 2026Haibo Jin, Peng Kuang, Xucheng Yu +3Repository-Level Code Generation Software Engineering
University of Illinois Urbana-Champaign, USA
E2E-SWE: Benchmarking LLMs on Building Working Codebases from Scratch
Sep 29, 2026Hantian Ding, Chloe Bi, Jiacheng Zhu +5Software Engineering Agents Repository-Level Code Generation
Meta Superintelligence Labs
Zero2Repo: Can Coding Agents Build Repositories from Scratch?
Sep 29, 2026Pei Yang, Tianyu Shi, Yuhang Yao +23Coding Agents AI Agent Benchmarks
Gradient Data · McGill University · Carnegie Mellon University +6
Do Coding Agents Reuse Existing Code or Reinvent the Wheel?
Sep 28, 2026Dongsheng Ma, Sizhe Wang, Xinyi Huang +5Coding Agents Code Generation
Peking University · Fudan University · Shanghai Jiao Tong University +3
FeatLens: Feature-Guided Dynamic Code Graph Construction and Retrieval for Repository-Level Code Generation
Sep 22, 2026Xutian Li, Bo Xiong, Yifeng Zhu +6Graph Retrieval-Augmented Generation Repository-Level Code Generation
Peking University
From Reading Code to Reading Spec: A Verified Layer for LLM-Driven Codebase Maintenance
Sep 6, 2026Xinhao Zhang, Jingjie Lu, Kunpeng Liu +1Automated Software Engineering Repository-Level Code Generation
Portland State University, USA · Clemson University, USA
PaperCompiler: Faithful Paper-to-Code Generation via Repository-Level Specification Compilation
Sep 2, 2026Yunhao Liu, Hong Phuc Pham, Jaehong YoonCode Generation Repository-Level Code Generation
NTU Singapore
Adaptive Critical Token-Aware Retrieval for Repository-Level Code Generation
Sep 1, 2026Kefeng Duan, Dewu Zheng, Yanlin Wang +8Retrieval-Augmented Generation Adaptive RAG
Sun Yat-sen University, Guangzhou 510006, P.R. China · Huawei Cloud Computing Technologies Co., Ltd.
WiseSpec: Requirements-Driven Agents for Code Generation
Sep 1, 2026Zhao TianRequirements Engineering Software Engineering Agents
School of Computer Software, Tianjin University Tianjin, China
Identifying Latent Declarative Representations of Code for Assisting Repository Migration
Aug 22, 2026Shraddha Surana, Ashwin Srinivasan, Michael BainCode Translation Repository-Level Code Generation
Department of Computer Science and Information Systems, BITS Pilani, K. K. Birla Goa Campus, NH 17B, Zuarinagar, 403726, Goa, India · School of Computer Science and Engineering, UNSW Sydney, Sydney, NSW, 2052, Australia
Repo0: Design-Driven Zero-to-All Code Generation
Aug 20, 2026Silin Chen, Haoyi Teng, Xiaodong Gu +5Code Generation Agentic Code Generation
Shanghai Jiao Tong University · Chongqing University
TDD-Agent: Test-Driven Reasoning for Code Generation
Aug 17, 2026Hongyue Yu, Kefan Li, Jiakun Li +4Code Generation Execution-Guided Code Generation
National College for Excellent Engineers, Beihang University · School of Computer Science and Engineering, Beihang University · Qingdao Research Institute and Hangzhou Innovation Institute, Beihang University +1
The Working Set of a Coding Agent: Coherence Debt in Repository-Scale Tasks
Aug 17, 2026Bardia Mohammadi, Lars Klein, Aman Chadha +2Repository-Level Code Generation LLM Agent Memory
Max Planck Institute for Software Systems · EPFL · Apple +1
Vero: Can AI Agents Build Formally Verified Software Repositories?
Aug 13, 2026Zhe Ye, Hantao Lou, Yuechun Sun +8Code Generation Formal Verification
1UC Berkeley · University of Chicago · 3California Institute of Technology +3
Persistent Recursive Worlds Enable Autonomous Software Evolution
Aug 11, 2026Beichen Huang, Zhenyu Liang, Bowen Zheng +1Automated Software Engineering Repository-Level Code Generation
Department of Data Science and Artificial Intelligence, The Hong Kong Polytechnic University Hong Kong SAR, China
Can Coding Agents Solve Repository-Level Issues with Rendered Code? An Exploratory Study of Visual Representations
Aug 10, 2026Weijie Liang, Yuanfeng Song, Xing Chen +3Automated Program Repair AI Coding Agents
The Hong Kong University of Science and Technology, Hong Kong, China · ByteDance, China
Effective and Efficient Context Retrieval via Partial Dependency Graph for Repository-Level Code Generation
Aug 3, 2026Zhongxin Liu, Zhonghao Jiang, Zhifan Ye +3Retrieval-Augmented Generation Graph-Based Retrieval
College of Computer Science and Technology and the State Key Laboratory of Blockchain and Data Security, Zhejiang University · School of Computer and Computing Science, Hangzhou City University · Faculty of Computing, Harbin Institute of Technology +1
Beyond "What to Retrieve": Uncertainty in Retrieval-Augmented Code Generation
Jul 27, 2026Chandan Kumar Sah, Li Zhang, Xiaoli LianRetrieval-Augmented Generation LLM Uncertainty Estimation
Beihang University, Beijing, China
The Patchwork Problem in LLM-Generated Code
Jul 9, 2026Viraaji Mothukuri, Reza M. PariziStatic Code Analysis LLM Reliability
Decentralized Science Lab, College of Computing and Software Engineering Kennesaw State University, GA, USA
ProjAgent: Procedural Similarity Retrieval for Repository-Level Code Generation
Jul 9, 2026QiHong Chen, Aaron Imani, Iftekhar AhmedCode Generation Agentic RAG
University of California, Irvine, USA
KAT-Coder-V2.5 Technical Report
Jul 6, 2026Bo Huang, Fengxiang Li, Hao Xu +50RL for Code Generation Software Engineering Agents
CodeTeam: An LLM-Powered Multi-Agent Framework for Repository-Level Code Generation
Jun 20, 2026Yifei Wang, Ruiyin Li, Peng Liang +4Repository-Level Code Generation Multi-Agent Coding
School of Computer Science, Wuhan University, China · School of Computer Science, Nanjing University of Science and Technology, China · School of Computer Science, Central China Normal University, China +2
JAMER: Project-Level Code Framework Dataset and Benchmark on Professional Game Engines
Jun 18, 2026Jianwen Sun, Chuanhao Li, Zizhen Li +5Benchmark Design Code Generation
1Alaya Studio · 2Nankai University, Tianjin, China · 3Shanghai Innovation Institute, Shanghai, China
When Retrieval Hurts Code Completion: A Diagnostic Study of Stale Repository Context
May 14, 2026Haojun Weng, Qianqian Yang, Hao Fu +2Retrieval-Augmented Generation Retrieval Robustness
Independent Researcher, California, USA · Independent Researcher, Beijing, China
BUILD-AND-FIND: An Effort-Aware Protocol for Evaluating Agent-Managed Codebases
May 7, 2026Jhen-Ke LinBenchmark Construction LLM Agent Evaluation
National Yang Ming Chiao Tung University
LLM-Assisted Repository-Level Generation with Structured Spec-Driven Engineering
May 4, 2026Shuzhao Feng, Boqi Chen, Brett H Meyer +1Code Generation Repository-Level Code Generation
McGill University · Montreal, Quebec, Canada · University of Ottawa +1

Oct 26, 2026J/K move · Enter open · S save

Skill Dreaming: Evolving Repository Skills through Pull Request Streams

Xinyuan Song, Hasibul Haque, Jay Somasundaram, Ryan Turner+2

Causal Dynamics Lab, USA

Automated Software Engineering LLM Agent Skill Learning Repository-Level Code Generation Continual Learning for LLM Agents

Open paper AskPDF arXiv

Large language model (LLM) agents increasingly rely on reusable skills for complex software engineering tasks, making skill libraries essential for repository development and maintenance. However, repository-specific skills must evolve with changing code structures, implementation conventions, and maintenance requirements. Pull requests (PRs) provide a natural source of such knowledge by recording how repositories evolve through implementation and review. Meanwhile, new PRs must follow repository-specific implementation patterns, modify related components consistently, and ensure that existing tests continue to pass while adding coverage for new functionality. This creates a mutually reinforcing cycle: PRs provide knowledge for updating skills, while these skills improve future PRs, maintain repositories, and support further skill evolution. To enable this co-evolution of PRs, repositories, and skills, we propose Skill Dreaming, a framework that learns from PR streams to guide continuous repository development and maintenance. Given a repository and its PR history, Skill Dreaming builds a repository-specific skill library by extracting executable knowledge from validated PRs and continuously refines the skills as new PR information becomes available. These skills guide coding agents not only in implementing requested features but also in maintaining associated tests, documentation, compatibility, and repository conventions. By integrating PR analysis, skill management, code execution, and continuous integration, Skill Dreaming provides an end-to-end workflow for applying learned skills to repository changes and incorporating subsequent PR outcomes into future skill updates.