Guardado en:
Detalles Bibliográficos
Autores principales: Li, Mengdi, Liu, Yuming, Wang, He, Xu, Zifeng, Zhang, Yuqing
Formato: Preprint
Publicado: 2026
Materias:
Acceso en línea:https://arxiv.org/abs/2603.15004
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866915865745686528
author Li, Mengdi
Liu, Yuming
Wang, He
Xu, Zifeng
Zhang, Yuqing
author_facet Li, Mengdi
Liu, Yuming
Wang, He
Xu, Zifeng
Zhang, Yuqing
contents 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.
format Preprint
id arxiv_https___arxiv_org_abs_2603_15004
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle TriFusion-LLM: Prior-Guided Multimodal Fusion with LLM Arbitration for Fine-grained Code Clone Detection
Li, Mengdi
Liu, Yuming
Wang, He
Xu, Zifeng
Zhang, Yuqing
Software Engineering
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.
title TriFusion-LLM: Prior-Guided Multimodal Fusion with LLM Arbitration for Fine-grained Code Clone Detection
topic Software Engineering
url https://arxiv.org/abs/2603.15004