Papers › GenCompareSum: a hybrid unsupervised summarization method using salience

GenCompareSum: a hybrid unsupervised summarization method using salience

1 May 2022BioNLP (ACL) 2022 5archive 2025-07-28

Jennifer Bishop, Qianqian Xie, Sophia Ananiadou

Text summarization (TS) is an important NLP task. Pre-trained Language Models (PLMs) have been used to improve the performance of TS. However, PLMs are limited by their need of labelled training data and by their attention mechanism, which often makes them unsuitable for use on long documents. To this end, we propose a hybrid, unsupervised, abstractive-extractive approach, in which we walk through a document, generating salient textual fragments representing its key points. We then select the most important sentences of the document by choosing the most similar sentences to the generated texts, calculated using BERTScore. We evaluate the efficacy of generating and using salient textual fragments to guide extractive summarization on documents from the biomedical and general scientific domains. We compare the performance between long and short documents using different generative text models, which are finetuned to generate relevant queries or document titles. We show that our hybrid approach out-performs existing unsupervised methods, as well as state-of-the-art supervised methods, despite not needing a vast amount of labelled training data.

PaperPDFCode

Code

jbshp/gencomparesum 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

Extractive SummarizationText Summarization

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Text Summarization Arxiv HEP-TH citation graph GenCompareSum ROUGE-1 39.96 #25 of 28 Archive leaderboard report
Text Summarization Arxiv HEP-TH citation graph GenCompareSum ROUGE-2 15.15 #25 of 28 Archive leaderboard report
Text Summarization Arxiv HEP-TH citation graph GenCompareSum ROUGE-L 36.19 #25 of 28 Archive leaderboard report
Text Summarization CORD-19 GenCompareSum ROUGE-1 41.02 #1 of 1 Archive leaderboard report
Text Summarization CORD-19 GenCompareSum ROUGE-2 13.79 #1 of 1 Archive leaderboard report
Text Summarization CORD-19 GenCompareSum ROUGE-L 37.25 #1 of 1 Archive leaderboard report
Text Summarization Pubmed GenCompareSum ROUGE-1 42.10 #24 of 29 Archive leaderboard report
Text Summarization Pubmed GenCompareSum ROUGE-2 16.51 #24 of 29 Archive leaderboard report
Text Summarization Pubmed GenCompareSum ROUGE-L 38.25 #24 of 29 Archive leaderboard report
Text Summarization S2ORC GenCompareSum ROUGE-1 43.39 #1 of 1 Archive leaderboard report
Text Summarization S2ORC GenCompareSum ROUGE-2 16.84 #1 of 1 Archive leaderboard report
Text Summarization S2ORC GenCompareSum ROUGE-L 39.82 #1 of 1 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

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