Document-aware Positional Encoding and Linguistic-guided Encoding for Abstractive Multi-document Summarization
This work addresses a specific bottleneck in multi-document summarization for NLP applications, offering incremental improvements over existing methods.
The paper tackled the challenge of capturing cross-document relations and linguistic structure in multi-document summarization by proposing document-aware positional encoding and linguistic-guided encoding integrated with Transformer architecture, resulting in high-quality generated summaries as shown in experiments.
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.