arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2212.07428cs.CLcs.LG
收藏

面向自然语言推理的语言信息多目标预训练

Towards Linguistically Informed Multi-Objective Pre-Training for Natural Language Inference

Fraunhofer IAIS(弗劳恩霍夫智能分析与信息系统研究所);Ruhr-Universität Bochum(波鸿鲁尔大学);Rheinische Friedrich-Wilhelms Universität Bonn(莱茵兰-弗里德里希-威廉姆斯波恩大学)

Maren Pielka, Svetlana Schmidt, Lisa Pucknat, Rafet Sifa

摘要:

我们引入了一种语言增强的Transformers预训练方法的组合。预训练的目标包括POS标注、基于语义知识图谱的同义词集预测和基于依存解析树的父类预测。与现有技术相比,我们的方法在自然语言推理任务上取得了有竞争力的结果。特别是对于较小的模型,该方法可以显著提高性能,强调智能预训练可以弥补更少的参数,并有助于构建更有效的模型。将POS标记和同义词集预测相结合会产生总体上最好的结果。

英文摘要:

We introduce a linguistically enhanced combination of pre-training methods for transformers. The pre-training objectives include POS-tagging, synset prediction based on semantic knowledge graphs, and parent prediction based on dependency parse trees. Our approach achieves competitive results on the Natural Language Inference task, compared to the state of the art. Specifically for smaller models, the method results in a significant performance boost, emphasizing the fact that intelligent pre-training can make up for fewer parameters and help building more efficient models. Combining POS-tagging and synset prediction yields the overall best results.

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。

↑