@inproceedings{c7a10ddd128d4c689c352befbe3c2413,
title = "Insert Commonsense Knowledge Through Semantics for Dialogue Generation",
abstract = "Language models such as GPT and BART have been widely applied to various language generation tasks, including dialogue generation. While some works have improved the generation performance of these models by incorporating external knowledge, the effective extraction of valuable knowledge from diverse sources and its incorporation into generation models require further investigation. To tackle this problem, we propose a method called Knowledge-enhanced Multi-turn Dialogue Model (KMDM) to extract commonsense knowledge from external knowledge graphs and inject the knowledge into the encoding and decoding processes globally and locally. We first extract sub-graphs according to the semantic correlations with contexts. Then we add a hierarchical graph attention layer to the decoder in order to acquire local information from sub-graphs. Experiments conducted on Wizard of Wikipedia and DailyDialog show the effectiveness of the proposed method.",
keywords = "Commonsense Knowledge Graph, Dialogue, Natural Language Generation",
author = "Siqi Hou and Dandan Song and Zhijing Wu and Xiechao Guo and Ziyi Yang",
note = "Publisher Copyright: {\textcopyright} The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024.; 17th International Conference on Knowledge Science, Engineering and Management, KSEM 2024 ; Conference date: 16-08-2024 Through 18-08-2024",
year = "2024",
doi = "10.1007/978-981-97-5495-3_23",
language = "English",
isbn = "9789819754946",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
publisher = "Springer Science and Business Media Deutschland GmbH",
pages = "305--317",
editor = "Cungeng Cao and Huajun Chen and Liang Zhao and Junaid Arshad and Yonghao Wang and Taufiq Asyhari",
booktitle = "Knowledge Science, Engineering and Management - 17th International Conference, KSEM 2024, Proceedings",
address = "Germany",
}