提出双向动态注意力网络，提升文本分类的上下文捕捉能力。
A Dual-Directional Context-Aware Test-Time Learning for Text Classification
- 结合双向时序建模与自注意力机制，动态加权关键输入片段。
- 在保持高效计算的同时，增强对长距离依赖的建模能力。
- 适合需要高精度和快速推理的文本分类场景。
文本分类将文本分配到预定义类别中。传统方法在处理复杂结构和长距离依赖方面存在困难。深度学习中的循环神经网络和Transformer模型虽提升了特征提取与上下文感知能力，但仍面临可解释性、效率与上下文范围之间的权衡。本文提出动态双向Elman注意力网络（DBEAN），融合双向时序建模与自注意力机制，能够动态加权关键输入片段，并在保持计算效率的同时增强上下文建模能力。
原文摘要 · Abstract (English)
Text classification assigns text to predefined categories. Traditional methods struggle with complex structures and long-range dependencies. Deep learning with recurrent neural networks and Transformer models has improved feature extraction and context awareness. However, these models still trade off interpretability, efficiency and contextual range. We propose the Dynamic Bidirectional Elman Attention Network (DBEAN). DBEAN combines bidirectional temporal modeling and self-attention. It dynamically weights critical input segments and preserves computational efficiency.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。