Multi-Scale Feature Fusion and Graph Neural Network Integration for Text Classification with Large Language Models
University of Michigan(密歇根大学);Georgia Institute of Technology(佐治亚理工学院);Dartmouth College(达特茅斯学院);University of Pennsylvania(宾夕法尼亚大学);University of Southern California(南加州大学)
This study investigates a hybrid method for text classification that integrates deep feature extraction from large language models, multi-scale fusion through feature pyramids, and structured modeling with graph neural networks to enhance performance in complex semantic contexts. First, the large language model captures contextual dependencies and deep semantic representations of the input text, providing a rich feature foundation for subsequent modeling. Then, based on multi-level feature representations, the feature pyramid mechanism effectively integrates semantic features of different scales, balancing global information and local details to construct hierarchical semantic expressions. Furthermore, the fused features are transformed into graph representations, and graph neural networks are employed to capture latent semantic relations and logical dependencies in the text, enabling comprehensive modeling of complex interactions among semantic units. On this basis, the readout and classification modules generate the final category predictions. The proposed method demonstrates significant advantages in robustness alignment experiments, outperforming existing models on ACC, F1-Score, AUC, and Precision, which verifies the effectiveness and stability of the framework. This study not only constructs an integrated framework that balances global and local information as well as semantics and structure, but also provides a new perspective for multi-scale feature fusion and structured semantic modeling in text classification tasks.
情感分析主要基于文本,近十年来发展迅速,引起了学术界和工业界的广泛关注。然而,现实世界中的信息通常来自多种形式,如音频和文本。因此,本文基于音频和文本,考虑多模态情感分析的任务,提出了一种新的融合策略,包括多特征融合和多模态融合,以提高音频-文本情感分析的准确性。我们称之为DFF-ATMF(Deep Feature Fusion - Audio and Text Modality Fusion)模型,它由两个并行的分支组成,即基于音频模态的分支和基于文本模态的分支。其核心机制是多特征向量和多通道注意的融合。在CMU-MOSI数据集和最近发布的CMU-MOSEI数据集上的实验,都是从YouTube上收集的情感分析,显示了我们的DFF-ATMF模型非常有竞争力的结果。此外,凭借注意力权重分布热图,我们还证明了使用DFF-ATMF学习的深度特征是互补的,并且是鲁棒的。令人惊讶的是,DFF-ATMF在IEMOCAP数据集上也取得了新的最先进的结果,表明所提出的融合策略对于多模态情感识别也具有良好的泛化能力。
A Deep Neural Information Fusion Architecture for Textual Network Embeddings
School of Data and Computer Science, Sun Yat-sen University(中山大学数据与计算机科学学院);School of Electronics and Information Technology, Sun Yat-sen University(中山大学电子与信息工程学院)