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 2009.04968cs.CLcs.LG
收藏

现代文本生成方法

Modern Methods for Text Generation

National Research University Higher School of Economics(国立研究型大学高等经济学院)

Dimas Munoz Montesinos

摘要:

合成文本生成具有挑战性,并且成功有限。最近,一种名为Transformers的新架构允许机器学习模型更好地理解顺序数据,例如翻译或摘要。BERT和GPT-2在其内核中使用了Transformers,在文本分类、翻译和NLI等任务中表现出了出色的性能。在这篇文章中,我们分析了这两种算法,并比较了它们在文本生成任务中的输出质量。

英文摘要:

Synthetic text generation is challenging and has limited success. Recently, a new architecture, called Transformers, allow machine learning models to understand better sequential data, such as translation or summarization. BERT and GPT-2, using Transformers in their cores, have shown a great performance in tasks such as text classification, translation and NLI tasks. In this article, we analyse both algorithms and compare their output quality in text generation tasks.

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。

↑