Retrieval-Augmented Code Generation: A Survey with Focus on Repository-Level Approaches

2025-10-06

AI Summary

This paper addresses core challenges in repository-level code generation (RLCG)—namely, difficulty in modeling long-range dependencies, cross-file semantic inconsistency, and weak global structural coherence. To this end, it introduces the first systematic, retrieval-augmented generation (RAG)-based unified analytical framework tailored for RLCG. The framework features a multi-granularity (function-/file-/repository-level) classification scheme that elucidates key mechanisms for cross-file dependency modeling and global consistency maintenance. It integrates context-aware generation, modular architecture design, and multi-source information retrieval to enhance both output quality and scalability. Furthermore, the paper comprehensively surveys prevailing datasets and evaluation benchmarks, identifying critical bottlenecks in dynamic context updating, semantic consistency assurance, and large-scale repository adaptation. Collectively, this work provides foundational theoretical insights and practical technical pathways for AI-driven software engineering.

Technology Category

• Natural Language Processing: Code Generation / Program Synthesis from Natural Language
• Machine Learning: Deep Generative Models & Autoencoders
• Cognitive Modeling & Cognitive Systems: Agent Architectures

Application Category

• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Social Networks and Social Media: Generative AI / large language models and their impact on social systems

Abstract

Recent advancements in large language models (LLMs) have substantially improved automated code generation. While function-level and file-level generation have achieved promising results, real-world software development typically requires reasoning across entire repositories. This gives rise to the challenging task of Repository-Level Code Generation (RLCG), where models must capture long-range dependencies, ensure global semantic consistency, and generate coherent code spanning multiple files or modules. To address these challenges, Retrieval-Augmented Generation (RAG) has emerged as a powerful paradigm that integrates external retrieval mechanisms with LLMs, enhancing context-awareness and scalability. In this survey, we provide a comprehensive review of research on Retrieval-Augmented Code Generation (RACG), with an emphasis on repository-level approaches. We categorize existing work along several dimensions, including generation strategies, retrieval modalities, model architectures, training paradigms, and evaluation protocols. Furthermore, we summarize widely used datasets and benchmarks, analyze current limitations, and outline key challenges and opportunities for future research. Our goal is to establish a unified analytical framework for understanding this rapidly evolving field and to inspire continued progress in AI-powered software engineering.

Problem

• Addressing repository-level code generation challenges across multiple files
• Enhancing context-awareness in code generation using retrieval-augmented approaches
• Improving global semantic consistency in large-scale software development

Innovation

• Retrieval-augmented generation enhances code generation
• Repository-level approaches address multi-file dependencies
• External retrieval mechanisms improve context-awareness and scalability