Topic-Guided Coherence Modeling for Sentence Ordering by Preserving Global and Local Information

Oh, Byungkook; Seo, Seungmin; Shin, Cheolheon; Jo, Eun-Ju; Lee, Kyong-Ho

doi:10.18653/v1/d19-1232

Cited by 20 publications

(19 citation statements)

References 31 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Assume that there are Q paragraphs in the training set Q = {(s, o)}. Following the existing ordering networks (Gong et al, 2016;Oh et al, 2019), the model is trained to maximize the coherence probability by minimizing the loss function as follows:…”

Section: Model Trainingmentioning

confidence: 99%

“…Following the existing work (Oh et al, 2019), we employ the three most commonly used metrics 5 in this task to assess the model performance: Accuracy (Acc): This metric calculates the ratio of sentences whose absolute positions are correctly predicted (Logeswaran et al, 2018). Perfect Match Ratio (PMR): It measures the percentage of the exactly matching orders across all the paragraphs: PMR= 1…”

Section: Evaluation Metricsmentioning

confidence: 99%

“…(2) Pointer network based models: HAN (Wang and Wan, 2019); LSTM+PtrNet (Gong et al, 2016); V-LSTM+PtrNet (Logeswaran et al, 2018); ATTOrderNet (Cui et al, 2018); SE-Graph (Yin et al, 2019); FUDecoder (Yin et al, 2020); TGCM (Oh et al, 2019).…”

Section: Baselinesmentioning

confidence: 99%

“…Previous studies (Oh et al, 2019;Yin et al, 2019) have mentioned that both the first and last sentences play crucial roles in a paragraph due to their special positions. Thus, we also report the performances of our models in correctly predicting these two sentences on arXiv and SIND datasets.…”

Section: Prediction Of First and Last Sentencesmentioning

confidence: 99%

“…It is able to capture the paragraphlevel contextual information for generating an ordered sequence with the highest coherence probability (Gong et al, 2016;Logeswaran et al, 2018;Cui et al, 2018;Yin et al, 2019). Further, HAN (Wang and Wan, 2019) and TGCM (Oh et al, 2019) introduce the attention mechanism (Vaswani et al, 2017), and FUDecoder (Yin et al, 2020) proposes pairwise ordering prediction modules to enhance the traditional pointer network.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

BERT-enhanced Relational Sentence Ordering Network

Cui¹,

Li²,

Zhang³

2020

Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP)

View full text Add to dashboard Cite

In this paper, we introduce a novel BERTenhanced Relational Sentence Ordering Network (referred to as BERSON) by leveraging BERT for capturing a better dependency relationship among sentences to enhance the coherence modeling for the entire paragraph. In particular, we develop a new Relational Pointer Decoder (referred as RPD) by incorporating the relative ordering information into the pointer network with a Deep Relational Module (referred as DRM), which utilizes BERT to exploit the deep semantic connection and relative ordering between sentences. This enables us to strengthen both local and global dependencies among sentences. Extensive evaluations are conducted on six public datasets. The experimental results demonstrate the effectiveness and promise of BERSON, showing a significant improvement over the state-of-the-art by a wide margin. * Corresponding author An unordered set of sentences Coherent paragraph 1 Dan was walking during the night.1 Dan was walking during the night. 3 They tried to steal his book bag.2 A group of thieves surrounded him. 4 A bystander noticed them.3 They tried to steal his book bag. 2 A group of thieves surrounded him. 4 A bystander noticed them. 5 But she continued to walk away.5 But she continued to walk away.

show abstract

Section: Model Trainingmentioning

confidence: 99%

Section: Evaluation Metricsmentioning

confidence: 99%

Section: Baselinesmentioning

confidence: 99%

Section: Prediction Of First and Last Sentencesmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

BERT-enhanced Relational Sentence Ordering Network

Cui¹,

Li²,

Zhang³

2020

Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP)

View full text Add to dashboard Cite

show abstract

An overview of sentence ordering task

Shi,

Zhang,

et al. 2024

Int J Data Sci Anal

View full text Add to dashboard Cite

The sentence ordering task aims to organize complex, unordered sentences into readable text. This improves accuracy, validity, and reliability in various natural language processing domains, including automatic text generation, text summarization, and machine translation. We begin by analyzing and summarizing the sentence ordering algorithm from two perspectives: the input data approach and the implementation technique approach. Based on the different ways of input data formats, they are classified into pointwise, pairwise, and listwise, and the advantages, disadvantages and representative algorithmic features of each are discussed. Based on the different implementation technologies, we classify them into sentence ordering algorithms based on learning to rank and deep learning, and the core ideas, typical algorithms and research progress of these two categories of methods were specifically explained. We summarize the datasets and evaluation metrics of currently commonly used sentence ordering tasks. Additionally, we analyze the problems and challenges of sentence ordering tasks and look forward to the future direction of this field.

show abstract