Abstract
Most current extractive summarization models generate summaries by selecting salient sentences. However, one of the problems with sentence-level extractive summarization is that there exists a gap between the human-written gold summary and the oracle sentence labels. In this paper, we propose to extract fact-level semantic units for better extractive summarization. We also introduce a hierarchical structure, which incorporates the multi-level of granularities of the textual information into the model. In addition, we incorporate our model with BERT using a hierarchical graph mask. This allows us to combine BERT’s ability in natural language understanding and the structural information without increasing the scale of the model. Experiments on the CNN/DaliyMail dataset show that our model achieves state-of-the-art results.- Anthology ID:
- 2020.coling-main.493
- Volume:
- Proceedings of the 28th International Conference on Computational Linguistics
- Month:
- December
- Year:
- 2020
- Address:
- Barcelona, Spain (Online)
- Editors:
- Donia Scott, Nuria Bel, Chengqing Zong
- Venue:
- COLING
- SIG:
- Publisher:
- International Committee on Computational Linguistics
- Note:
- Pages:
- 5629–5639
- Language:
- URL:
- https://aclanthology.org/2020.coling-main.493
- DOI:
- 10.18653/v1/2020.coling-main.493
- Cite (ACL):
- Ruifeng Yuan, Zili Wang, and Wenjie Li. 2020. Fact-level Extractive Summarization with Hierarchical Graph Mask on BERT. In Proceedings of the 28th International Conference on Computational Linguistics, pages 5629–5639, Barcelona, Spain (Online). International Committee on Computational Linguistics.
- Cite (Informal):
- Fact-level Extractive Summarization with Hierarchical Graph Mask on BERT (Yuan et al., COLING 2020)
- PDF:
- https://preview.aclanthology.org/nschneid-patch-2/2020.coling-main.493.pdf
- Code
- Ruifeng-paper/FactExsum-coling2020