@article{Tong2025, 
author = {Xin Tong and Jingya Wang and Ying Yang and Tian Peng and Hanming Zhai and Guangming Ling},
title = {LEGF-DST: LLMs-Enhanced Graph-Fusion Dual-Stream Transformer for Fine-Grained Chinese Malicious SMS Detection},
year = {2025},
journal = {Computers, Materials & Continua},
volume = {82},
number = {2},
pages = {1901-1924},
keywords = {Transformers, malicious SMS, multi-task learning, large language models},
url = {https://www.sciopen.com/article/10.32604/cmc.2024.059018},
doi = {10.32604/cmc.2024.059018},
abstract = {With the widespread use of SMS (Short Message Service), the proliferation of malicious SMS has emerged as a pressing societal issue. While deep learning-based text classifiers offer promise, they often exhibit suboptimal performance in fine-grained detection tasks, primarily due to imbalanced datasets and insufficient model representation capabilities. To address this challenge, this paper proposes an LLMs-enhanced graph fusion dual-stream Transformer model for fine-grained Chinese malicious SMS detection. During the data processing stage, Large Language Models (LLMs) are employed for data augmentation, mitigating dataset imbalance. In the data input stage, both word-level and character-level features are utilized as model inputs, enhancing the richness of features and preventing information loss. A dual-stream Transformer serves as the backbone network in the learning representation stage, complemented by a graph-based feature fusion mechanism. At the output stage, both supervised classification cross-entropy loss and supervised contrastive learning loss are used as multi-task optimization objectives, further enhancing the model’s feature representation. Experimental results demonstrate that the proposed method significantly outperforms baselines on a publicly available Chinese malicious SMS dataset.}
}