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 2210.07431cs.CLcs.LG
收藏

基于PCFG的自然语言接口提高了受控文本生成的泛化能力

PCFG-based Natural Language Interface Improves Generalization for Controlled Text Generation

Johns Hopkins University(约翰斯·霍普金斯大学);MIT(麻省理工学院);University of Washington(华盛顿大学)

Jingyu Zhang, James Glass, Tianxing He

摘要:

现有的工作控制文本生成(CTG)假设一个控制接口的分类属性。在这项工作中,我们提出了一个自然语言(NL)的接口,我们工艺PCFG嵌入控制属性到自然语言命令,并提出了现有的CTG模型,以命令作为输入的变体。在我们的实验中,我们设计了定制的设置来测试模型的泛化能力。我们发现我们的基于PCFG的命令生成方法是有效的处理看不见的命令相比,固定设置的模板,我们提出的NL模型可以有效地推广到看不见的属性,NL接口启用的新功能,以及看不见的属性组合。有趣的是,我们发现,简单的条件生成方法,增强了我们提出的NL接口,是一个强大的基线,在这些具有挑战性的设置。

英文摘要:

Existing work on controlled text generation (CTG) assumes a control interface of categorical attributes. In this work, we propose a natural language (NL) interface, where we craft a PCFG to embed the control attributes into natural language commands, and propose variants of existing CTG models that take commands as input. In our experiments, we design tailored setups to test model's generalization abilities. We find our PCFG-based command generation approach is effective for handling unseen commands compared to fix-set templates; our proposed NL models can effectively generalize to unseen attributes, a new ability enabled by the NL interface, as well as unseen attribute combinations. Interestingly, we discover that the simple conditional generation approach, enhanced with our proposed NL interface, is a strong baseline in those challenging settings.

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。

↑