Papers › Abstractive Summarization Guided by Latent Hierarchical Document Structure

Abstractive Summarization Guided by Latent Hierarchical Document Structure

17 Nov 2022arXiv:2211.09458archive 2025-07-28

Yifu Qiu, Shay B. Cohen

Sequential abstractive neural summarizers often do not use the underlying structure in the input article or dependencies between the input sentences. This structure is essential to integrate and consolidate information from different parts of the text. To address this shortcoming, we propose a hierarchy-aware graph neural network (HierGNN) which captures such dependencies through three main steps: 1) learning a hierarchical document structure through a latent structure tree learned by a sparse matrix-tree computation; 2) propagating sentence information over this structure using a novel message-passing node propagation mechanism to identify salient information; 3) using graph-level attention to concentrate the decoder on salient information. Experiments confirm HierGNN improves strong sequence models such as BART, with a 0.55 and 0.75 margin in average ROUGE-1/2/L for CNN/DM and XSum. Further human evaluation demonstrates that summaries produced by our model are more relevant and less redundant than the baselines, into which HierGNN is incorporated. We also find HierGNN synthesizes summaries by fusing multiple source sentences more, rather than compressing a single source sentence, and that it processes long inputs more effectively.

PaperPDFCode

In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

Code

yfqiu-nlp/hiergnn officialmentioned in paperpytorch report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

Abstractive Text SummarizationDecoderGraph Neural NetworkSentence

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

Methods

AdamAttentionBARTBPEDense ConnectionsDropoutGraph Neural NetworkLayer NormalizationLinear LayerMulti-Head AttentionResidual ConnectionSoftmax

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections