Event-Keyed Summarization

William Gantt, Alexander Martin, Pavlo Kuchmiichuk, Aaron Steven White
2/10/2024
cs.CLcs.AIcs.LG

No official code found. Best implementation: wgantt/eks (3 stars, MIT, last commit 2024-02-13).

Abstract

We introduce event-keyed summarization (EKS), a novel task that marries traditional summarization and document-level event extraction, with the goal of generating a contextualized summary for a specific event, given a document and an extracted event structure. We introduce a dataset for this task, MUCSUM, consisting of summaries of all events in the classic MUC-4 dataset, along with a set of baselines that comprises both pretrained LM standards in the summarization literature, as well as larger frontier models. We show that ablations that reduce EKS to traditional summarization or structure-to-text yield inferior summaries of target events and that MUCSUM is a robust benchmark for this task. Lastly, we conduct a human evaluation of both reference and model summaries, and provide some detailed analysis of the results.

View on arXivView PDF

Code Implementations(1)

wgantt/eksCommunity100%
MIT

Event-Keyed Summarization

30PythonFeb 10, 20242 years agoMIT

Similar Papers

Cross-Document Event-Keyed Summarization

87% match1 repo2024

EventSum: A Large-Scale Event-Centric Summarization Dataset for Chinese Multi-News Documents

84% match2024

Event-based evaluation of abstractive news summarization

81% match2025

Scaling Multi-Document Event Summarization: Evaluating Compression vs. Full-Text Approaches

81% match1 repo2025

Sentence-wise Speech Summarization: Task, Datasets, and End-to-End Modeling with LM Knowledge Distillation

79% match2024

From News to Summaries: Building a Hungarian Corpus for Extractive and Abstractive Summarization

79% match1 repo112024

Cite this paper

@article{gantt2024eventkeyed,
  title  = {Event-Keyed Summarization},
  author = {William Gantt and Alexander Martin and Pavlo Kuchmiichuk and Aaron Steven White},
  year   = {2024},
  eprint = {2402.06973},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2402.06973v1}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2402.06973.svg)](https://codeofpaper.com/paper/2402.06973)

Discussion