arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2106.03337cs.CL
收藏

摘要固定对话生成

Summary Grounded Conversation Generation

IBM Research AI(IBM研究院人工智能部门)

Chulaka Gunasekara, Guy Feigenblat, Benjamin Sznajder, Sachindra Joshi, David Konopnicki

摘要:

近年来,许多会话数据集已经使用众包构建。然而,数据汇集过程可能很耗时,并对确保数据质量提出了许多挑战。由于语言生成有很大的改善,近年来与预先训练的语言模型的进步,我们研究如何利用这些模型来生成整个会话,只给出一个会话的摘要作为输入。我们探索了三种方法来生成摘要接地会话,并使用自动测量和人类判断来评估生成的会话。我们还表明,会话摘要的准确性可以通过增加生成的会话摘要数据集来提高。

英文摘要:

Many conversation datasets have been constructed in the recent years using crowdsourcing. However, the data collection process can be time consuming and presents many challenges to ensure data quality. Since language generation has improved immensely in recent years with the advancement of pre-trained language models, we investigate how such models can be utilized to generate entire conversations, given only a summary of a conversation as the input. We explore three approaches to generate summary grounded conversations, and evaluate the generated conversations using automatic measures and human judgements. We also show that the accuracy of conversation summarization can be improved by augmenting a conversation summarization dataset with generated conversations.

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。

↑