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 1909.01528cs.CLcs.LG
收藏

使用实体配置文件生成引用表达式

Referring Expression Generation Using Entity Profiles

McGill University(麦吉尔大学);MILA(蒙特利尔学习算法研究所)

Meng Cao, Jackie Chi Kit Cheung

摘要:

引用表达式生成(REG)是生成上下文适当的实体引用的任务。现有REG系统的局限性在于它们依赖于实体特定的监督训练,这意味着它们不能处理在训练期间看不到的实体。在这项研究中,我们以两种方式解决这个问题。首先,我们提出了任务设置,其中我们专门测试REG系统的能力,以推广到在训练过程中看不到的实体。其次,我们提出了一个基于配置文件的深度神经网络模型ProfileREG,它对实体的本地上下文和外部配置文件进行编码,以生成参考实现。我们的模型通过学习在生成代词、从固定词汇表生成或从配置文件中复制单词之间进行选择来生成令牌。我们在WebNLG数据集的三个不同部分上评估了我们的模型,并根据自动和人工评估表明,它在所有设置中都优于竞争基线。

英文摘要:

Referring Expression Generation (REG) is the task of generating contextually appropriate references to entities. A limitation of existing REG systems is that they rely on entity-specific supervised training, which means that they cannot handle entities not seen during training. In this study, we address this in two ways. First, we propose task setups in which we specifically test a REG system's ability to generalize to entities not seen during training. Second, we propose a profile-based deep neural network model, ProfileREG, which encodes both the local context and an external profile of the entity to generate reference realizations. Our model generates tokens by learning to choose between generating pronouns, generating from a fixed vocabulary, or copying a word from the profile. We evaluate our model on three different splits of the WebNLG dataset, and show that it outperforms competitive baselines in all settings according to automatic and human evaluations.

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。

↑