Problems in Current Text Simplification Research: New Data Can Help

By Wei Xu, Chris Callison-Burch, Courtney Napoles. Transactions of the Association for Computational Linguistics, 2015

496 citations recorded by the source corpus.

Simple Wikipedia has dominated simplification research in the past 5 years. In this opinion paper, we argue that focusing on Wikipedia limits simplification research. We back up our arguments with corpus analysis and by highlighting statements that other researchers have made in the simplification literature. We introduce a new simplification dataset that is a significant improvement over Simple Wikipedia, and present a novel quantitative-comparative approach to study the quality of simplification data resources.

Subfields: Artificial Intelligence, Communication, History and Philosophy of Science

View Paper: https://www.semanticscholar.org/paper/2b8da013966c0c5e020ebc842d49d8ed166c8783

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 Communication

In History and Philosophy of Science

Citations in the corpus

Decontextualization: Making Sentences Stand-Alone (2021, background): “(3)Research on text simplification (Xu et al., 2015; Bingel et al., 2018) also shows how target output depends on the expected audience.”
Paraphrasing evades detectors of AI-generated text, but retrieval is an effective defense (2023, methodology): “Our work is also related to efforts in text simpliﬁcation to go beyond a sentence, by collecting relevant datasets [Xu et al., 2015, Devaraj et al., 2021] and building unsupervised algorithms [Laban et al., 2021].”
PPDB 2.0: Better paraphrase ranking, fine-grained entailment relations, word embeddings, and style classification (2015, background): “These classiﬁcations could be useful for natural language generation tasks like text simpliﬁcation (Xu et al., 2015).”
PowerTransformer: Unsupervised Controllable Revision for Biased Language Correction (2020, methodology): “Controllable Debiasing is a new formalization of the unsupervised stylistic rewriting task, contrasting with supervised approaches which beneﬁt from parallel corpora (e.g., Xu et al., 2012, 2015; Rao and Tetreault, 2018; Pryzant et al., 2020).”
Neural CRF Model for Sentence Alignment in Text Simplification (2020, methodology): “Previous work (Xu et al., 2015) has shown that Wikipedia is much noisier than the Newsela corpus.”
Assessing Relative Sentence Complexity using an Incremental CCG Parser (2016, methodology): “We can explore the usefulness of our system on other datasets like On-eStopEnglish (OSE) corpus (Vajjala and Meurers, 2016) or the dataset from Xu et al. (2015).”
Returning the N to NLP: Towards Contextually Personalized Classification Models (2020, background): “hand, the ability to understand it is subjective, and some users may appreciate lexical simplification (Xu et al., 2015).”
Sentence Simplification with Deep Reinforcement Learning (2017, methodology): “Xu et al. (2015b) provide multiple aligned complex-simple pairs within each article.”
ASSET: A Dataset for Tuning and Evaluation of Sentence Simplification Models with Multiple Rewriting Transformations (2020, methodology): “More diverse simplifications are available in the Newsela corpus (Xu et al., 2015), a dataset of 1,130 news articles that were each manually simplified”
MUSS: Multilingual Unsupervised Sentence Simplification by Mining Paraphrases (2022, unknown): “We use two labeled datasets: Wik-iLarge (Zhang and Lapata, 2017) and Newsela (Xu et al., 2015).”
Split and Rephrase (2017, methodology): “Finally, sentence (or text) simpliﬁcation aims to produce a text that is easier to understand (Siddharthan et al., 2004; Zhu et al., 2010; Woodsend and Lapata, 2011; Wubben et al., 2012; Narayan and Gardent, 2014; Xu et al., 2015; Narayan and Gardent, 2016; Zhang and La-pata, 2017).”
Optimizing Statistical Machine Translation for Text Simplification (2016, background): “However, recent studies (Xu et al., 2015; Amancio and Specia, 2014; Hwang et al., 2015; ˇStajner et al., 2015) showed that the parallel Wikipedia simpliﬁ-cation corpus contains a large proportion of inadequate (not much simpler) or inaccurate (not aligned or only partially aligned) simpliﬁcations.”
Evaluating Factuality in Text Simplification (2022, methodology): “We could not successfully reproduce their results for Newsela. reported both Wikilarge and Newsela outputs, the relative frequency of deletion errors between the two datasets appears to be preserved in model outputs, though for the RNN models errors are milder on Newsela and amplified on Wikilarge.”
Equity Beyond Bias in Language Technologies for Education (2019, methodology): “NLP is also used in related tasks like the measurement of readability (Aluisio et al., 2010; Vajjala and Meurers, 2012), and generation of simpliﬁed texts to differentiate homework based on student ability (Xu et al., 2015).”
Reformulating Unsupervised Style Transfer as Paraphrase Generation (2020, background): “…been criticized for its limited real-world applications: Pang (2019) argue that se-1 mantic preservation is critical for author obfuscation (Shetty et al., 2018), data augmentation (Xie et al., 2019; Kaushik et al., 2020), text simpliﬁca-tion (Xu et al., 2015), writing assistance (Heidorn, 2000).”
Submodular Optimization-based Diverse Paraphrasing and its Effectiveness in Data Augmentation (2019, background): “Diversity in paraphrase generation finds applications in text simplification (Nisioi et al., 2017; Xu et al., 2015), document summarization (Li et al.”
Why Attention is Not Explanation: Surgical Intervention and Causal Reasoning about Neural Models (2020, background): “This approach resulted in a variety of visualizations and other attempts at model explanation being explored based, in part or in whole, on attention weights (Xu et al., 2015a; Mullenbach et al., 2018; Yang et al., 2019).”