The paper introduces SaraCoder, a retrieval‑augmented framework that improves knowledge density for repository‑level code completion by pruning redundant context and selectively injecting cross‑file dependency definitions within a fixed token budget. It employs a Hierarchical Feature Optimization module to reduce an over‑complete candidate pool in four stages, freeing tokens for an External‑Aware Identifier Disambiguator. Experiments on CrossCodeEval and RepoEval‑Updated demonstrate that SaraCoder outperforms strong RAG baselines such as GraphCoder and Repocoder across multiple code LLMs without increasing token usage, and achieves synergistic gains when combined with other cross‑file methods.
By Xiaohan Chen, Xiang Chen, Yu Tian, Piji Li, Sheng-Jun Huang, Songcan Chen
Code Comprehensor is a hybrid retrieval‑reasoning framework designed to aid understanding of large codebases. It combines a code dependency callgraph, CodeScribe—which generates hierarchical documentation using code‑aware LLMs—and CodeContrast, a CLIP‑style dual‑encoder that learns aligned text‑code embeddings for fast, semantically precise retrieval. Experiments on OpenCV and a 5M‑LOC proprietary repository show that CodeScribe‑derived supervision improves cross‑modal retrieval accuracy, and LLM‑based evaluation reports over 90% faithfulness.
By Sujoy Roychowdhury, Joseph H. R. Isaac, Jiangning Gao, Martin Klitte, Diarmuid Corcoran
arXiv:2607. 08691v1 Announce Type: cross Abstract: Repository-level code generation requires implementing target functions while accounting for complex cross-file dependencies and project-specific conventions.
By QiHong Chen, Aaron Imani, Iftekhar Ahmed
arXiv:2607. 24882v1 Announce Type: cross Abstract: Modern coding agents are usually evaluated by whether they eventually produce a correct patch, but patch generation depends on an earlier context-acquisition stage: finding the repository files needed for the task.
By Bowen Qin, Yi Xie
arXiv:2607. 24884v1 Announce Type: cross Abstract: Repository-level code generation relies on heterogeneous evidence whose relevance, compatibility, and completeness are inherently uncertain.
By Chandan Kumar Sah, Xiaoli Lian, Li Zhang
arXiv:2606.22906v2 Announce Type: replace-cross
Abstract: Large language models have shown strong performance on software engineering (SE) tasks, yet understanding large industrial repositories remai...
By Jiawei He, Weisong Sun, Mengyu Shi, Jie Jia, Tong Bian, Xikai Yang, Dong Sun
arXiv:2601. 19697v2 Announce Type: replace-cross Abstract: Repository-level code completion remains a challenging task for existing code large language models (code LLMs) due to their limited understanding of repository-specific context and domain knowledge.
By Tianyue Jiang, Yanli Wang, Yanlin Wang, Daya Guo, Ensheng Shi, Yuchi Ma, Jiachi Chen, Zibin Zheng
arXiv:2610.09079v1 Announce Type: cross
Abstract: Large language models equipped with development environments have moved code generation toward repository-scale construction, yet building complete r...
By Haibo Jin, Peng Kuang, Xucheng Yu, Jerry Wang, Dehao Wu, Haohan Wang
arXiv:2505. 13353v5 Announce Type: replace-cross Abstract: Large language models (LLMs) are increasingly deployed for understanding large codebases, but whether they understand operational semantics of long code context or rely on pattern matching shortcuts remains unclear.
By Adam \v{S}torek, Mukur Gupta, Samira Hajizadeh, Prashast Srivastava, Suman Jana
arXiv:2609.16936v1 Announce Type: cross
Abstract: Large language model (LLM)-powered coding agents have made rapid progress in automating software engineering tasks, yet repository-level issue resolu...
By Yunxiang Zhang, Haiquan Wang, JiaWei Guo, Hanyang Xia, Yan Chen, Tong Chen, Zhang Zhiwei, Junchen Ye
arXiv:2609. 31176v1 Announce Type: new Abstract: Repository-level issue localization aims to identify and rank the files and functions relevant to resolving a reported issue.
By Yunxiang Wei, Zhenyu Lei, Jundong Li
arXiv:2607. 01916v1 Announce Type: new Abstract: Large language model agents can repair real repository issues, but they often spend large context budgets on whole-file reads, broad searches, and long terminal outputs where useful evidence is mixed with irrelevant code and logs.
By Chiwang Luk, Matin Mohammad Najafi, Zhifeng Jia, Wei Yang, Xiuchang Li, Jinwei Zhu, Yang Ren, Lei Chen, Gao Cong