arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2005.00084cs.CL
收藏

神经元控制的神经元参数生成

Aspect-Controlled Neural Argument Generation

Ubiquitous Knowledge Processing Lab (UKP-TUDA)(泛在知识处理实验室(UKP-TUDA));Technische Universität Darmstadt(达姆施塔特工业大学)

Benjamin Schiller, Johannes Daxenberger, Iryna Gurevych

摘要:

我们依靠日常生活中的论点来表达我们的观点,并将其建立在证据的基础上,从而使它们更有说服力。然而,寻找和形成论点可能具有挑战性。在这项工作中,我们训练一个语言模型的参数生成,可以控制在一个细粒度的水平,以产生一个给定的主题,立场和方面的论据级别的参数。我们将参数方面检测定义为一种必要的方法,以允许这种细粒度控制和众包一个带有5,032个方面注释的参数的数据集。我们的评估表明,我们的生成模型能够生成高质量的,特定于方面的参数。此外,这些参数可以用于通过数据增强来提高姿态检测模型的性能,并生成反参数。我们发布所有数据集和代码来微调语言模型。

英文摘要:

We rely on arguments in our daily lives to deliver our opinions and base them on evidence, making them more convincing in turn. However, finding and formulating arguments can be challenging. In this work, we train a language model for argument generation that can be controlled on a fine-grained level to generate sentence-level arguments for a given topic, stance, and aspect. We define argument aspect detection as a necessary method to allow this fine-granular control and crowdsource a dataset with 5,032 arguments annotated with aspects. Our evaluation shows that our generation model is able to generate high-quality, aspect-specific arguments. Moreover, these arguments can be used to improve the performance of stance detection models via data augmentation and to generate counter-arguments. We publish all datasets and code to fine-tune the language model.

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。

↑