natural-language-processing-11

Extractive Text Summarization

3260 papers • 126 benchmarks • 313 datasets

Given a document, selecting a subset of the words or sentences which best represents a summary of the document.

(Image credit: Papersgraph)

Benchmarks

These leaderboards are used to track progress in extractive-text-summarization-11

Trend

Dataset

Best Model

Actions

CNN / Daily Mail

DebateSum

GovReport

Libraries

i

Use these libraries to find extractive-text-summarization-11 models and implementations

HHousen/TransformerSum

3 papers 429

Datasets

Subtasks

Reader-Aware Summarization

Most implemented papers

Get To The Point: Summarization with Pointer-Generator Networks

Peter J. Liu, Christopher D. Manning, A. See•Fri Mar 31 2017

A novel architecture that augments the standard sequence-to-sequence attentional model in two orthogonal ways, using a hybrid pointer-generator network that can copy words from the source text via pointing, which aids accurate reproduction of information, while retaining the ability to produce novel words through the generator.

4307

Content

DUC 2004 Task 1

DUC 2004

0

Paper Graph

Text Summarization with Pretrained Encoders

Mirella Lapata, Yang Liu•Wed Jul 31 2019

This paper introduces a novel document-level encoder based on BERT which is able to express the semantics of a document and obtain representations for its sentences and proposes a new fine-tuning schedule which adopts different optimizers for the encoder and the decoder as a means of alleviating the mismatch between the two.

1584 0

Paper Graph

Efficient Attention: Attention with Linear Complexities

Hongsheng Li, Mingyuan Zhang, Haiyu Zhao, Shuai Yi, Zhuoran Shen•Mon Dec 03 2018

A novel efficient attention mechanism equivalent to dot-product attention but with substantially less memory and computational costs is proposed, which allows more widespread and flexible integration of attention modules into a network, which leads to better accuracies.

674 0

Paper Graph

Fine-tune BERT for Extractive Summarization

Yang Liu•Sun Mar 24 2019

BERTSUM, a simple variant of BERT, for extractive summarization, is described, which is the state of the art on the CNN/Dailymail dataset, outperforming the previous best-performed system by 1.65 on ROUGE-L.

520 0

Paper Graph

Leveraging BERT for Extractive Text Summarization on Lectures

Derek Miller•Thu Jun 06 2019

This paper reports on the project called Lecture Summarization Service, a python based RESTful service that utilizes the BERT model for text embeddings and KMeans clustering to identify sentences closes to the centroid for summary selection.

266 0

Paper Graph

A Neural Attention Model for Abstractive Sentence Summarization

S. Chopra, J. Weston, Alexander M. Rush•Tue Sep 01 2015

This work proposes a fully data-driven approach to abstractive sentence summarization by utilizing a local attention-based model that generates each word of the summary conditioned on the input sentence.

2807 0

Paper Graph

AREDSUM: Adaptive Redundancy-Aware Iterative Sentence Ranking for Extractive Document Summarization

Asli Celikyilmaz, W. Bruce Croft, Rahul Jha, Keping Bi•Sun Apr 12 2020

Two adaptive learning models are presented: AREDSUM-SEQ that jointly considers salience and novelty during sentence selection; and a two-step AREDsUM-CTX that scores salience first, then learns to balancesalience and redundancy, enabling the measurement of the impact of each aspect.

23 0

Paper Graph

DebateSum: A large-scale argument mining and summarization dataset

Allen Roush, Arvind Balaji•Fri Nov 13 2020

The DebateSum dataset, which consists of 187,386 unique pieces of evidence with corresponding argument and extractive summaries, is presented and a search engine for this dataset is presented, utilized extensively by members of the National Speech and Debate Association today.

39 0

Paper Graph

Centroid-based Text Summarization through Compositionality of Word Embeddings

Gaetano Rossiello, Pierpaolo Basile, G. Semeraro•Sat Dec 31 2016

A centroid-based method for text summarization that exploits the compositional capabilities of word embeddings and achieves good performance even in comparison to more complex deep learning models.

136 0

Paper Graph

TAP-DLND 1.0 : A Corpus for Document Level Novelty Detection

Asif Ekbal, P. Bhattacharyya, Tirthankar Ghosal, A. Salam, Swati Tiwary•Mon Feb 19 2018

This work creates a resource for benchmarking the techniques for document level novelty detection via event-specific crawling of news documents across several domains in a periodic manner and releases the annotated corpus with necessary statistics.

24 0

Paper Graph

Adding a benchmark result helps the community track progress.