AST-Enhanced or AST-Overloaded? The Surprising Impact of Hybrid Graph Representations on Code Clone Detection

TL;DR

This study empirically evaluates how hybrid graph representations like CFG, DFG, and FA-AST affect GNN-based code clone detection, revealing that certain enrichments improve accuracy while others may hinder performance, with GMN showing strong results even with basic ASTs.

Contribution

It provides a comprehensive empirical comparison of hybrid AST-based graph representations in GNNs for code clone detection, highlighting the effectiveness of different structures and models.

Findings

Hybrid representations like CFG and DFG improve GNN accuracy.

FA-AST can introduce complexity that harms performance.

GMN outperforms other models with standard ASTs.

Abstract

As one of the most detrimental code smells, code clones significantly increase software maintenance costs and heighten vulnerability risks, making their detection a critical challenge in software engineering. Abstract Syntax Trees (ASTs) dominate deep learning-based code clone detection due to their precise syntactic structure representation, but they inherently lack semantic depth. Recent studies address this by enriching AST-based representations with semantic graphs, such as Control Flow Graphs (CFGs) and Data Flow Graphs (DFGs). However, the effectiveness of various enriched AST-based representations and their compatibility with different graph-based machine learning techniques remains an open question, warranting further investigation to unlock their full potential in addressing the complexities of code clone detection. In this paper, we present a comprehensive empirical study to rigorously evaluate the effectiveness of AST-based hybrid graph representations in Graph Neural Network (GNN)-based code clone detection. We systematically compare various hybrid representations ((CFG, DFG, Flow-Augmented ASTs (FA-AST)) across multiple GNN architectures. Our experiments reveal that hybrid representations impact GNNs differently: while AST+CFG+DFG consistently enhances accuracy for convolution- and attention-based models (Graph Convolutional Networks (GCN), Graph Attention Networks (GAT)), FA-AST frequently introduces structural complexity that harms performance. Notably, GMN outperforms others even with standard AST representations, highlighting its superior cross-code similarity detection and reducing the need for enriched structures.