BanditSum: Extractive Summarization as a Contextual Bandit

By Yue Dong, Yikang Shen, Eric Crawford, Herke Van Hoof, Jackie Chi Kit Cheung. 2018

189 citations recorded by the source corpus.

In this work, we propose a novel method for training neural networks to perform single-document extractive summarization without heuristically-generated extractive labels. We call our approach BanditSum as it treats extractive summarization as a contextual bandit (CB) problem, where the model receives a document to summarize (the context), and chooses a sequence of sentences to include in the summary (the action). A policy gradient reinforcement learning algorithm is used to train the model to select sequences of sentences that maximize ROUGE score. We perform a series of experiments demonstrating that BanditSum is able to achieve ROUGE scores that are better than or comparable to the state-of-the-art for extractive summarization, and converges using significantly fewer update steps than competing approaches. In addition, we show empirically that BanditSum performs significantly better than competing approaches when good summary sentences appear late in the source document.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition, Natural Language Processing

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Computer Vision and Pattern Recognition

In Natural Language Processing

Citations in the corpus

Fine-tune BERT for Extractive Summarization (2019, background): “…many neural models have been proposed for extractive summarization recently (Cheng and Lapata, 2016; Nallapati et al., 2017; Narayan et al., 2018; Dong et al., 2018; Zhang et al., 2018; Zhou et al., 2018), the improvement on automatic metrics like ROUGE has reached a bottleneck due to the…”
Learning to summarize from human feedback (2020, background): “There is also a rich literature on using RL to optimize automatic metrics for NLP tasks, such as ROUGE for summarization [50, 65, 45, 15, 19], BLEU for translation [50, 66, 1, 43], and other domains [61, 27, 26].”
Better Rewards Yield Better Summaries: Learning to Summarise Without References (2019, methodology): “8 (Dong et al., 2018) R-1,2,L 41.”
Text Summarization with Pretrained Encoders (2019, methodology): “…proposed techniques aim to enhance summarization performance via copying mechanisms (Gu et al., 2016; See et al., 2017; Nallap-ati et al., 2017), reinforcement learning (Narayan et al., 2018b; Paulus et al., 2018; Dong et al., 2018), and multiple communicating encoders (Ce-likyilmaz et al., 2018).”
Evaluating the Factual Consistency of Abstractive Text Summarization (2019, background): “2018; Pasunuru and Bansal, 2018), including reinforcement learning techniques (Kryściński et al., 2018; Dong et al., 2018; Wu and Hu, 2018).”
Neural Text Summarization: A Critical Evaluation (2019, methodology): “…et al., 2018; Pasunuru and Bansal, 2018; Kry´sci´nski et al., 2018), reinforcement learning strategies (Paulus et al., 2017; Narayan et al., 2018b; Dong et al., 2018; Wu and Hu, 2018), and hybrid extractive-abstractive models (Liu et al., 2018; Hsu et al., 2018; Gehrmann et al., 2018; Chen and…”
BARTScore: Evaluating Generated Text as Text Generation (2021, neutral): “Extractive Systems E1: BanditSum [11]; E2: Refresh [49]; E3: NeuSum [81]; E4: LSTM-PN-RL [80]; E5: BERT-TF-SL [80]; E6: BERT-TF-PN [80]; E7: BERT-LSTM-PN-RL [80]; E8: BERT-LSTM-PN [80]; E9: HeterGraph [69]; E10: MatchSum [79].”
HIBERT: Document Level Pre-training of Hierarchical Bidirectional Transformers for Document Summarization (2019, methodology): “The architecture is widely adopted in recent neural ex-tractive models and is extended with reinforcement learning (Narayan et al., 2018; Dong et al., 2018), latent variable models (Zhang et al., 2018), joint scoring (Zhou et al., 2018) and iterative document representation (Chen et al., 2018).”
Optimizing the Factual Correctness of a Summary: A Study of Summarizing Radiology Reports (2019, methodology): “To study whether the improvements in the factual correctness scores lead to improvement in summarization quality under expert judgment, we run a comparative human evaluation following previous work (Chen and Bansal, 2018; Dong et al., 2018; Zhang et al., 2018).”
RefSum: Refactoring Neural Summarization (2021, unknown): “6 In details, for §5.7 we use the following systems: pointer-generator+coverage (See et al., 2017), REFRESH (Narayan et al., 2018b), fastAbsRL-rank (Chen and Bansal, 2018), CNN-LSTM-BiClassiﬁer (Kedzie et al., 2018), CNN-Transformer-BiClassiﬁer (Zhong et al., 2019), CNN-Transformer-Pointer (Zhong et”
SummEval: Re-evaluating Summarization Evaluation (2020, background): “M2 - BanditSum (Dong et al., 2018) treats extractive summarization as a contextual bandit problem where the document is the context and the sequence of sentences to include in the summary is the action.”
Re-evaluating Evaluation in Text Summarization (2020, unknown): “We use CNN-LSTM-BiClassiﬁer (CLSTM-SL; Kedzie et al. (2018)), Latent (Zhang et al., 2018), Ban-ditSum (Dong et al., 2018), REFRESH (Narayan et al., 2018), NeuSum (Zhou et al., 2018), HIBERT (Zhang et al., 2019b), Bert-Sum-Ext (Liu and Lapata, 2019a), CNN-Transformer-BiClassiﬁer (CTrans-SL; Zhong et…”
Extractive Summarization as Text Matching (2020, methodology): “BANDITSUM (Dong et al., 2018) 41.”
GUIR @ LongSumm 2020: Learning to Generate Long Summaries from Scientific Documents (2020, background): “…ways: 1) Extractive: where the system ﬁnds the salient sentences within the source and concate-nates them to form the summary (Zhou et al., 2018; Dong et al., 2018; Zhang et al., 2018; Narayan et al., 2018; Liu and Lapata, 2019; Xu et al., 2020); and 2) Abstractive: where the model conducts…”
Discourse-Aware Neural Extractive Text Summarization (2019, methodology): “98 BanditSum (Dong et al., 2018) 41.”
Dialect Diversity in Text Summarization on Twitter (2020, background): “Beyond unsupervised approaches, supervised techniques classify whether a sentence is important or not [45, 21, 56, 33].”