Retrieve, Rerank and Rewrite: Soft Template Based Neural Summarization

Cao, Zhigang; Li, Wenjie; Li, Sujian; Wei, Furu

doi:10.18653/v1/p18-1015

Cited by 180 publications

(161 citation statements)

References 18 publications

(23 reference statements)

Supporting

Mentioning

161

Contrasting

Order By: Relevance

“…Recent summarization research has put special emphasis on faithfulness to the original text. Cao et al (2018a) use seq-to-seq models to rewrite templates that are prone to including irrelevant entities. Incorporating additional information into a seq-to-seq model, such as entailment and dependency structure, has proven successful (Li et al, 2018;Song et al, 2018).…”

Section: Related Workmentioning

confidence: 99%

Analyzing Sentence Fusion in Abstractive Summarization

Lebanoff¹,

Muchovej²,

Dernoncourt³

et al. 2019

Proceedings of the 2nd Workshop on New Frontiers in Summarization

View full text Add to dashboard Cite

While recent work in abstractive summarization has resulted in higher scores in automatic metrics, there is little understanding on how these systems combine information taken from multiple document sentences. In this paper, we analyze the outputs of five state-of-the-art abstractive summarizers, focusing on summary sentences that are formed by sentence fusion. We ask assessors to judge the grammaticality, faithfulness, and method of fusion for summary sentences. Our analysis reveals that system sentences are mostly grammatical, but often fail to remain faithful to the original article.

show abstract

Section: Related Workmentioning

confidence: 99%

Analyzing Sentence Fusion in Abstractive Summarization

Lebanoff¹,

Muchovej²,

Dernoncourt³

et al. 2019

Proceedings of the 2nd Workshop on New Frontiers in Summarization

View full text Add to dashboard Cite

show abstract

“…We study a sample of 26 recent papers from major ACL conferences and outline the trends of manual evaluation in summarization in Table 1. From 26 papers, 11 papers (e.g., See et al, 2017;Kedzie et al, 2018;Cao et al, 2018) did not conduct any manual evaluation. Following the Document Understanding Conference (DUC, Dang, 2005), a majority of work has focused on evaluating the content and the linguistic quality of summaries (Nenkova, 2005).…”

Section: Literature Reviewmentioning

confidence: 99%

HighRES: Highlight-based Reference-less Evaluation of Summarization

Hardy¹,

Narayan²,

Vlachos³

2019

Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics

View full text Add to dashboard Cite

There has been substantial progress in summarization research enabled by the availability of novel, often large-scale, datasets and recent advances on neural network-based approaches. However, manual evaluation of the system generated summaries is inconsistent due to the difficulty the task poses to human non-expert readers. To address this issue, we propose a novel approach for manual evaluation, HIGHlight-based Reference-less Evaluation of Summarization (HIGHRES), in which summaries are assessed by multiple annotators against the source document via manually highlighted salient content in the latter. Thus summary assessment on the source document by human judges is facilitated, while the highlights can be used for evaluating multiple systems. To validate our approach we employ crowd-workers to augment with highlights a recently proposed dataset and compare two state-of-the-art systems. We demonstrate that HIGHRES improves inter-annotator agreement in comparison to using the source document directly, while they help emphasize differences among systems that would be ignored under other evaluation approaches. 1

show abstract

“…Guu et al (2018) also attempted to encode humanwritten sentences to improve neural text generation. Similar to our work, Cao et al (2018a) proposed to retrieve a related summary from the training set as soft template to assist with the summarization. However, their approach tends to oversimplify the role of the template, by directly concatenating a template after the source article encoding.…”

Section: Related Workmentioning

confidence: 95%

“…This module starts with a standard information retrieval library 2 to retrieve a small set of candidates for fine-grained filtering as Cao et al (2018a). To do that, all non-alphabetic characters (e.g., dates) are removed to eliminate their influence on article matching.…”

Section: Retrievementioning

confidence: 99%

BiSET: Bi-directional Selective Encoding with Template for Abstractive Summarization

Wang¹,

Quan²,

Wang³

2019

Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics

View full text Add to dashboard Cite

The success of neural summarization models stems from the meticulous encodings of source articles. To overcome the impediments of limited and sometimes noisy training data, one promising direction is to make better use of the available training data by applying filters during summarization. In this paper, we propose a novel Bi-directional Selective Encoding with Template (BiSET) model, which leverages template discovered from training data to softly select key information from each source article to guide its summarization process. Extensive experiments on a standard summarization dataset were conducted and the results show that the template-equipped BiSET model manages to improve the summarization performance significantly with a new state of the art.

show abstract

Retrieve, Rerank and Rewrite: Soft Template Based Neural Summarization

Cited by 180 publications

References 18 publications

Analyzing Sentence Fusion in Abstractive Summarization

Analyzing Sentence Fusion in Abstractive Summarization

HighRES: Highlight-based Reference-less Evaluation of Summarization

BiSET: Bi-directional Selective Encoding with Template for Abstractive Summarization

Contact Info

Product

Resources

About