Fine-tune BERT for Extractive Summarization

Yang Liu
3/25/2019
cs.CL

Official code: nlpyang/BertSum (1,504 stars, Apache-2.0, last commit 2022-01-11).

Abstract

BERT, a pre-trained Transformer model, has achieved ground-breaking performance on multiple NLP tasks. In this paper, we describe BERTSUM, a simple variant of BERT, for extractive summarization. Our system is the state of the art on the CNN/Dailymail dataset, outperforming the previous best-performed system by 1.65 on ROUGE-L. The codes to reproduce our results are available at https://github.com/nlpyang/BertSum

View on arXivView PDF

Code Implementations(1)

nlpyang/BertSumOfficial100%
Apache-2.0

Code for paper Fine-tune BERT for Extractive Summarization

1,504410PythonMar 25, 20194 years agoApache-2.0

Similar Papers

Text Summarization with Pretrained Encoders

87% match1 repo1.3k2019

Self-Supervised Learning for Contextualized Extractive Summarization

84% match1 repo392019

Extractive Summarization via ChatGPT for Faithful Summary Generation

83% match2023

AST-MHSA : Code Summarization using Multi-Head Self-Attention

83% match2023

DiffuSum: Generation Enhanced Extractive Summarization with Diffusion

83% match1 repo202023

PromptSum: Parameter-Efficient Controllable Abstractive Summarization

83% match2023

Cite this paper

@article{liu2019finetune,
  title  = {Fine-tune BERT for Extractive Summarization},
  author = {Yang Liu},
  year   = {2019},
  eprint = {1903.10318},
  archivePrefix = {arXiv},
  url    = {https://arxiv.org/abs/1903.10318}
}

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/1903.10318.svg)](https://codeofpaper.com/paper/1903.10318)

Discussion