A Bag of Tricks for Dialogue Summarization

A Bag of Tricks for Dialogue Summarization
复制标题

对话总结的技巧包

DOI:
10.18653/v1/2021.emnlp-main.631
复制
发表时间:
2021
期刊:
ArXiv
影响因子:
--
通讯作者:
K. McKeown
K. McKeown
中科院分区:
--
文献类型:
--
作者:
Muhammad Khalifa;Miguel Ballesteros;K. McKeown

文献摘要

参考文献

被引文献

相似文献

与新闻或科技文章摘要相比,对话摘要有其独特的挑战。在这项工作中,我们探讨了任务的四个不同的挑战:处理和区分属于多个扬声器,否定理解,推理的情况下,和非正式的语言理解的对话部分。使用预训练的序列到序列语言模型,我们探索了说话人姓名替换,否定范围突出显示,相关任务的多任务学习以及域内数据的预训练。我们的实验表明,我们提出的技术确实提高了摘要性能,优于强基线。
Dialogue summarization comes with its own peculiar challenges as opposed to news or scientific articles summarization. In this work, we explore four different challenges of the task: handling and differentiating parts of the dialogue belonging to multiple speakers, negation understanding, reasoning about the situation, and informal language understanding. Using a pretrained sequence-to-sequence language model, we explore speaker name substitution, negation scope highlighting, multi-task learning with relevant tasks, and pretraining on in-domain data. Our experiments show that our proposed techniques indeed improve summarization performance, outperforming strong baselines.
DOI: 10.18653/v1/p19-1210
发表时间: 2019-07
期刊: --
影响因子: --
作者:
Manling Li;Lingyu Zhang;Heng Ji;R. Radke
通讯作者: Manling Li;Lingyu Zhang;Heng Ji;R. Radke