TriFusion-LLM: Prior-Guided Multimodal Fusion with LLM Arbitration for Fine-grained Code Clone Detection

TL;DR

This paper introduces TriFusion-LLM, a multimodal fusion framework that combines heuristic priors, structural signals, and semantic embeddings with LLM arbitration to improve fine-grained code clone detection accuracy efficiently.

Contribution

It presents a novel fusion framework that integrates multiple representations and uses LLM arbitration to enhance fine-grained code clone detection beyond binary classification.

Findings

Macro-F1 score increased from 0.695 to 0.875 on BigCloneBench.

Selective arbitration of 0.2% high-uncertainty samples improves Macro-F1 by 0.3.

Fusion of structural, statistical, and semantic signals enhances clone type discrimination.

Abstract

Code clone detection (CCD) supports software maintenance, refactoring, and security analysis. Although pre-trained models capture code semantics, most work reduces CCD to binary classification, overlooking the heterogeneity of clone types and the seven fine-grained categories in BigCloneBench. We present Full Model, a multimodal fusion framework that jointly integrates heuristic similarity priors from classical machine learning, structural signals from abstract syntax trees (ASTs), and deep semantic embeddings from CodeBERT into a single predictor. By fusing structural, statistical, and semantic representations, Full Model improves discrimination among fine-grained clone types while keeping inference cost practical. On the seven-class BigCloneBench benchmark, Full Model raises Macro-F1 from 0.695 to 0.875. Ablation studies show that using the primary model's probability distribution as a prior to guide selective arbitration by a large language model (LLM) substantially outperforms blind reclassification; arbitrating only ~0.2% of high-uncertainty samples yields an additional 0.3 absolute Macro-F1 gain. Overall, Full Model achieves an effective performance-cost trade-off for fine-grained CCD and offers a practical solution for large-scale industrial deployment.