arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2212.09682cs.CL
收藏

希伯来语NLP的多语言序列到序列模型

Multilingual Sequence-to-Sequence Models for Hebrew NLP

Google Research(谷歌研究院)

Matan Eyal, Hila Noga, Roee Aharoni, Idan Szpektor, Reut Tsarfaty

摘要:

最近的工作将NLP的进展归因于大语言模型(LM),其模型大小增加,预训练数据量大。尽管如此,与其他语言的LM相比,目前最先进的希伯来语LM参数化不足,训练不足。此外,以前关于预训练的希伯来语LM的工作主要集中在编码器模型上。虽然仅编码器架构有利于分类任务,但考虑到希伯来语在形态上丰富的性质,它无法很好地满足子词预测任务,例如命名实体识别。在本文中,我们认为,序列到序列生成架构更适合LLM的情况下,形态丰富的语言(MRL),如希伯来语。我们证明,通过将Hebrew NLP管道中的任务转换为文本到文本任务,我们可以利用强大的多语言,预训练的序列到序列模型作为mT5,从而消除了对专门的,基于词素的,单独微调的解码器的需求。使用这种方法,我们的实验表明,在现有的希伯来语NLP基准上,与以前发表的结果相比,有了很大的改进。这些结果表明,多语言序列到序列模型为MRL的NLP提供了一个有前途的构建模块。

英文摘要:

Recent work attributes progress in NLP to large language models (LMs) with increased model size and large quantities of pretraining data. Despite this, current state-of-the-art LMs for Hebrew are both under-parameterized and under-trained compared to LMs in other languages. Additionally, previous work on pretrained Hebrew LMs focused on encoder-only models. While the encoder-only architecture is beneficial for classification tasks, it does not cater well for sub-word prediction tasks, such as Named Entity Recognition, when considering the morphologically rich nature of Hebrew. In this paper we argue that sequence-to-sequence generative architectures are more suitable for LLMs in the case of morphologically rich languages (MRLs) such as Hebrew. We demonstrate that by casting tasks in the Hebrew NLP pipeline as text-to-text tasks, we can leverage powerful multilingual, pretrained sequence-to-sequence models as mT5, eliminating the need for a specialized, morpheme-based, separately fine-tuned decoder. Using this approach, our experiments show substantial improvements over previously published results on existing Hebrew NLP benchmarks. These results suggest that multilingual sequence-to-sequence models present a promising building block for NLP for MRLs.

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。

↑