arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2501.09451cs.CL
收藏

基于弧向量化与注意力优化的图依存句法分析的规模扩展

Scaling Graph-Based Dependency Parsing with Arc Vectorization and Attention-Based Refinement

Université Sorbonne Paris Nord(巴黎第十三大学(索邦巴黎北大学));CNRS(法国国家科学研究中心);Laboratoire d’Informatique de Paris Nord(巴黎北信息学实验室)

Nicolas Floquet, Joseph Le Roux, Nadi Tomeh, Thierry Charnois

AI总结:

本研究提出一种基于弧向量化与注意力优化的图依存句法分析新架构,统一弧评分与标注流程,通过Transformer层模拟高阶依存关系,在PTB和UD数据集上实现了准确率与效率的双重提升。

摘要:

我们提出一种用于图依存句法分析的新型架构,该架构显式构建向量,弧和标签的评分均基于这些向量计算。我们的方法将弧评分与标注统一到单个网络中,解决了标准双流水线方法的关键局限,减少了由信息瓶颈和参数共享缺失导致的可扩展性问题。此外,我们的架构通过Transformer层克服了弧交互有限的问题,可高效模拟高阶依存关系。在PTB和UD上的实验表明,我们的模型在准确率和效率上均优于当前最优的句法分析器。

英文摘要:

We propose a novel architecture for graph-based dependency parsing that explicitly constructs vectors, from which both arcs and labels are scored. Our method addresses key limitations of the standard two-pipeline approach by unifying arc scoring and labeling into a single network, reducing scalability issues caused by the information bottleneck and lack of parameter sharing. Additionally, our architecture overcomes limited arc interactions with transformer layers to efficiently simulate higher-order dependencies. Experiments on PTB and UD show that our model outperforms state-of-the-art parsers in both accuracy and efficiency.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑