Document-aware Positional Encoding and Linguistic-guided Encoding for Abstractive Multi-document Summarization
Ma, Congbo, Zhang, Wei Emma, Pitawela, Pitawelayalage Dasun Dileepa, Qu, Yutong, Zhuang, Haojie, Wang, Hu
–arXiv.org Artificial Intelligence
One key challenge in multi-document summarization is to capture the relations among input documents that distinguish between single document summarization (SDS) and multi-document summarization (MDS). Few existing MDS works address this issue. One effective way is to encode document positional information to assist models in capturing cross-document relations. However, existing MDS models, such as Transformer-based models, only consider token-level positional information. Moreover, these models fail to capture sentences' linguistic structure, which inevitably causes confusions in the generated summaries. Therefore, in this paper, we propose document-aware positional encoding and linguistic-guided encoding that can be fused with Transformer architecture for MDS. For document-aware positional encoding, we introduce a general protocol to guide the selection of document encoding functions. For linguistic-guided encoding, we propose to embed syntactic dependency relations into the dependency relation mask with a simple but effective non-linear encoding learner for feature learning. Extensive experiments show the proposed model can generate summaries with high quality.
arXiv.org Artificial Intelligence
Sep-13-2022
- Country:
- Africa > Ethiopia
- Addis Ababa > Addis Ababa (0.04)
- Asia
- Europe
- Belgium > Brussels-Capital Region
- Brussels (0.04)
- France (0.04)
- Italy > Tuscany
- Florence (0.05)
- Netherlands (0.04)
- Spain > Catalonia
- Barcelona Province > Barcelona (0.04)
- Switzerland > Geneva
- Geneva (0.04)
- Belgium > Brussels-Capital Region
- North America
- Canada > British Columbia
- United States
- California
- Merced County > Los Banos (0.04)
- San Francisco County > San Francisco (0.05)
- Minnesota > Hennepin County
- Minneapolis (0.14)
- New York (0.04)
- California
- Oceania > Australia
- South Australia > Adelaide (0.05)
- Pacific Ocean > North Pacific Ocean
- San Francisco Bay (0.04)
- Africa > Ethiopia
- Genre:
- Research Report (1.00)
- Technology: