AlignCoder: Aligning Retrieval with Target Intent for Repository-Level Code Completion

TL;DR

AlignCoder enhances repository-level code completion by improving retrieval accuracy through query enhancement and reinforcement learning, significantly outperforming baselines across multiple benchmarks and models.

Contribution

It introduces a novel query enhancement mechanism and reinforcement learning-based retriever training for better retrieval in code completion.

Findings

Achieves 18.1% improvement in EM score on CrossCodeEval

Demonstrates high generalizability across models and languages

Outperforms existing retrieval methods in code completion tasks

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. While retrieval-augmented generation (RAG) approaches have shown promise by retrieving relevant code snippets as cross-file context, they suffer from two fundamental problems: misalignment between the query and the target code in the retrieval process, and the inability of existing retrieval methods to effectively utilize the inference information. To address these challenges, we propose AlignCoder, a repository-level code completion framework that introduces a query enhancement mechanism and a reinforcement learning based retriever training method. Our approach generates multiple candidate completions to construct an enhanced query that bridges the semantic gap between the initial query and the target code. Additionally, we employ reinforcement learning to train an AlignRetriever that learns to leverage inference information in the enhanced query for more accurate retrieval. We evaluate AlignCoder on two widely-used benchmarks (CrossCodeEval and RepoEval) across five backbone code LLMs, demonstrating an 18.1% improvement in EM score compared to baselines on the CrossCodeEval benchmark. The results show that our framework achieves superior performance and exhibits high generalizability across various code LLMs and programming languages.