# Discourse-Aware Unsupervised Summarization of Long Scientific Documents

arXiv: https://arxiv.org/abs/2005.00513 · HTML: https://mlsanity.com/paper/2005.00513/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 1 May 2020.

## Built on by (most cited first, 4 of 4)

- [Efficient Attentions for Long Document Summarization](https://mlsanity.com/paper/2104.02112/) (2021)
- [Bringing Structure into Summaries: a Faceted Summarization Dataset for Long Scientific Documents](https://mlsanity.com/paper/2106.00130/) (2021)
- [Scientific Paper Extractive Summarization Enhanced by Citation Graphs](https://mlsanity.com/paper/2212.04214/) (2022)
- [SumHiS: Extractive Summarization Exploiting Hidden Structure](https://mlsanity.com/paper/2406.08215/) (2024)

## Abstract (arXiv)

We propose an unsupervised graph-based ranking model for extractive summarization of long scientific documents. Our method assumes a two-level hierarchical graph representation of the source document, and exploits asymmetrical positional cues to determine sentence importance. Results on the PubMed and arXiv datasets show that our approach outperforms strong unsupervised baselines by wide margins in automatic metrics and human evaluation. In addition, it achieves performance comparable to many state-of-the-art supervised approaches which are trained on hundreds of thousands of examples. These results suggest that patterns in the discourse structure are a strong signal for determining importance in scientific articles.