ReCode: Improving LLM-based Code Repair with Fine-Grained Retrieval-Augmented Generation

2025-09-02

AI Summary

Existing LLM-based code repair methods suffer from high training/inference overhead and low retrieval accuracy in RAG due to coarse-grained code embeddings. This paper proposes ReCode, a fine-grained retrieval-augmented in-context learning framework for efficient and accurate code repair. Its core contributions are: (1) algorithm-aware retrieval—leveraging a lightweight algorithm classifier to predict defect types and dynamically prune the retrieval space; and (2) a modular dual-encoder architecture that separately models code structure and natural language semantics to enable fine-grained cross-modal matching. We evaluate ReCode on the real-world user-defect dataset RACodeBench and competitive programming benchmarks. Results show that ReCode significantly reduces inference cost while outperforming state-of-the-art methods in repair accuracy.

Technology Category

• Natural Language Processing: Code Generation / Program Synthesis from Natural Language
• Planning, Routing, and Scheduling: Replanning and Plan Repair
• Machine Learning: Large Multimodal Models (LMMs)

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
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking

Abstract

Recent advances in large language models (LLMs) have demonstrated impressive capabilities in code-related tasks, such as code generation and automated program repair. Despite their promising performance, most existing approaches for code repair suffer from high training costs or computationally expensive inference. Retrieval-augmented generation (RAG), with its efficient in-context learning paradigm, offers a more scalable alternative. However, conventional retrieval strategies, which are often based on holistic code-text embeddings, fail to capture the structural intricacies of code, resulting in suboptimal retrieval quality. To address the above limitations, we propose ReCode, a fine-grained retrieval-augmented in-context learning framework designed for accurate and efficient code repair. Specifically, ReCode introduces two key innovations: (1) an algorithm-aware retrieval strategy that narrows the search space using preliminary algorithm type predictions; and (2) a modular dual-encoder architecture that separately processes code and textual inputs, enabling fine-grained semantic matching between input and retrieved contexts. Furthermore, we propose RACodeBench, a new benchmark constructed from real-world user-submitted buggy code, which addresses the limitations of synthetic benchmarks and supports realistic evaluation. Experimental results on RACodeBench and competitive programming datasets demonstrate that ReCode achieves higher repair accuracy with significantly reduced inference cost, highlighting its practical value for real-world code repair scenarios.

Problem

• Improving code repair accuracy with fine-grained retrieval
• Reducing high inference costs in LLM-based code repair
• Addressing structural code intricacies in retrieval strategies

Innovation

• Fine-grained retrieval-augmented generation framework
• Algorithm-aware retrieval strategy with type predictions
• Modular dual-encoder architecture for semantic matching