GNN-Coder: Boosting Semantic Code Retrieval with Combined GNNs and Transformer

TL;DR

GNN-Coder introduces a novel GNN-Transformer framework that leverages AST structures and a new graph pooling method to significantly improve semantic code retrieval accuracy, especially for complex code snippets.

Contribution

This paper pioneers the integration of GNNs with Transformers for code retrieval, proposing a new graph pooling technique and a metric for embedding uniformity to enhance performance.

Findings

Achieves 1-10% improvement in MRR on CSN dataset.

Attains 20% gain in zero-shot performance on CosQA.

Demonstrates superior ability to distinguish code snippets using new metrics.

Abstract

Code retrieval is a crucial component in modern software development, particularly in large-scale projects. However, existing approaches relying on sequence-based models often fail to fully exploit the structural dependencies inherent in code, leading to suboptimal retrieval performance, particularly with structurally complex code fragments. In this paper, we introduce GNN-Coder, a novel framework based on Graph Neural Network (GNN) to utilize Abstract Syntax Tree (AST). We make the first attempt to study how GNN-integrated Transformer can promote the development of semantic retrieval tasks by capturing the structural and semantic features of code. We further propose an innovative graph pooling method tailored for AST, utilizing the number of child nodes as a key feature to highlight the intrinsic topological relationships within the AST. This design effectively integrates both sequential and hierarchical representations, enhancing the model's ability to capture code structure and semantics. Additionally, we introduce the Mean Angular Margin (MAM), a novel metric for quantifying the uniformity of code embedding distributions, providing a standardized measure of feature separability. The proposed method achieves a lower MAM, indicating a more discriminative feature representation. This underscores GNN-Coder's superior ability to distinguish between code snippets, thereby enhancing retrieval accuracy. Experimental results show that GNN-Coder significantly boosts retrieval performance, with a 1\%-10\% improvement in MRR on the CSN dataset, and a notable 20\% gain in zero-shot performance on the CosQA dataset.