EMNLP Findings2024
Edit-Constrained Decoding for Sentence Simplification
Tatsuya Zetsu, Yuki Arase, Tomoyuki Kajiwara
Abstract
We propose edit operation based lexically constrained decoding for sentence simplification. In sentence simplification, lexical paraphrasing is one of the primary procedures for rewriting complex sentences into simpler correspondences. While previous studies have confirmed the efficacy of lexically constrained decoding on this task, their constraints can be loose and may lead to sub-optimal generation. We address this problem by designing constraints that replicate the edit operations conducted in simplification and defining stricter satisfaction conditions. Our experiments indicate that the proposed method consistently outperforms the previous studies on three English simplification corpora commonly used in this task.
Ask Lune
Ask anything. The answer quotes the paper.
Lune reads all of it and the papers around it, then shows you the passages behind every answer.
Builds on19
- BERTScore: Evaluating Text Generation with BERTTianyi Zhang, Varsha Kishore, Felix Wu, Kilian Q. Weinberger et al.ICLR 2020 · 8,443 citations
- BART: Denoising Sequence-to-Sequence Pre-training for Natural Language Generation, Translation, and ComprehensionMike Lewis, Yinhan Liu, Naman Goyal, Marjan Ghazvininejad et al.ACL 2020 · 1,224 citations
- Neural CRF Model for Sentence Alignment in Text SimplificationChao Jiang, Mounica Maddela, Wuwei Lan, Yang Zhong et al.ACL 2020 · 103 citations
- FELIX: Flexible Text Editing Through Tagging and InsertionJonathan Mallinson, Aliaksei Severyn, Eric Malmi, Guillermo GarridoEMNLP Findings 2020 · 59 citations
- Iterative Edit-Based Unsupervised Sentence SimplificationDhruv Kumar, Lili Mou, Lukasz Golab, Olga VechtomovaACL 2020 · 58 citations
Related papers
- GRS: Combining Generation and Revision in Unsupervised Sentence SimplificationMohammad Dehghan, Dhruv Kumar, Lukasz GolabACL Findings 2022 · 10 citations
- ASSET: A Dataset for Tuning and Evaluation of Sentence Simplification Models with Multiple Rewriting TransformationsFernando Alva-Manchego, Louis Martin, Antoine Bordes, Carolina Scarton et al.ACL 2020 · 12 citations
- A Non-Autoregressive Edit-Based Approach to Controllable Text SimplificationSweta Agrawal, Weijia Xu, Marine CarpuatACL Findings 2021 · 23 citations
- Discourse Level Factors for Sentence Deletion in Text SimplificationYang Zhong, Chao Jiang, Wei Xu, Junyi Jessy LiAAAI 2020 · 57 citations
- ParaLS: Lexical Substitution via Pretrained ParaphraserJipeng Qiang, Kang Liu, Yun Li, Yunhao Yuan et al.ACL 2023 · 9 citations
