Towards Improving Abstractive Summarization via Entailment Generation

Ramakanth Pasunuru, Han Guo, Mohit Bansal


Abstract
Abstractive summarization, the task of rewriting and compressing a document into a short summary, has achieved considerable success with neural sequence-to-sequence models. However, these models can still benefit from stronger natural language inference skills, since a correct summary is logically entailed by the input document, i.e., it should not contain any contradictory or unrelated information. We incorporate such knowledge into an abstractive summarization model via multi-task learning, where we share its decoder parameters with those of an entailment generation model. We achieve promising initial improvements based on multiple metrics and datasets (including a test-only setting). The domain mismatch between the entailment (captions) and summarization (news) datasets suggests that the model is learning some domain-agnostic inference skills.
Anthology ID:
W17-4504
Volume:
Proceedings of the Workshop on New Frontiers in Summarization
Month:
September
Year:
2017
Address:
Copenhagen, Denmark
Venue:
WS
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
27–32
Language:
URL:
https://www.aclweb.org/anthology/W17-4504
DOI:
10.18653/v1/W17-4504
Bib Export formats:
BibTeX MODS XML EndNote
PDF:
http://aclanthology.lst.uni-saarland.de/W17-4504.pdf