ABCD: A Graph Framework to Convert Complex Sentences to a Covering Set of Simple Sentences

Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing, pages 3919–3931 August 1–6, 2021. ©2021 Association for Computational Linguistics 3919 ABCD: A Graph Framework to Convert Complex Sentences to a Covering Set of Simple Sentences Yanjun Gao and Ting-Hao (Kenneth) Huang and Rebecca J. Passonneau Pennsylvania State University {yug125,txh710,rjp49}@psu.edu Abstract Atomic clauses are fundamental text units for understanding complex sentences. Identify- ing the atomic sentences within complex sen- tences is important for applications such as summarization, argument mining, discourse analysis, discourse parsing, and question an- swering. Previous work mainly relies on rule- based methods dependent on parsing. We pro- pose a new task to decompose each complex sentence into simple sentences derived from the tensed clauses in the source, and a novel problem formulation as a graph edit task. Our neural model learns to Accept, Break, Copy or Drop elements of a graph that combines word adjacency and grammatical dependencies. The full processing pipeline includes modules for graph construction, graph editing, and sen- tence generation from the output graph. We introduce DeSSE, a new dataset designed to train and evaluate complex sentence decompo- sition, and MinWiki, a subset of MinWikiSplit. ABCD achieves comparable performance as two parsing baselines on MinWiki. On DeSSE, which has a more even balance of complex sen- tence types, our model achieves higher accu- racy on the number of atomic sentences than an encoder-decoder baseline. Results include a detailed error analysis. 1 Introduction Atomic clauses are fundamental text units for understanding complex sentences. The ability to decompose complex sentences facilitates research that aims to identify, rank or relate distinct pred- ications, such as content selection in summariza- tion (Fang et al., 2016; Peyrard and Eckle-Kohler, 2017), labeling argumentative discourse units in argument mining (Jo et al., 2019) or elementary discourse units in discourse analysis (Mann and Thompson, 1986; Burstein et al., 1998; Demir et al., 2010), or extracting atomic propositions for ques- tion answering (Pyatkin et al., 2020). In this work, Orig Sokuhi was born in Fujian and was ordained at 17. SS1 Sokuhi was born in Fujian. SS2 Sokuhi was ordained at 17. Figure 1: Example of a complex sentence (Orig) rewritten as two simple sentences (SS1, SS2). Underlined words in the source are preserved in the same order in the two outputs, the conjunction and (red font) is dropped, and the subject Sokuhi (blue font) is copied to the second simple sentence. we propose a new task to decompose complex sen- tences into a covering set of simple sentences, with one simple output sentence per tensed clause in the source sentence. We focus on tensed clauses rather than other constituents because they are syn- tactically and semantically more prominent, thus more essential in downstream tasks like argument mining, summarization, and question answering. The complex sentence decomposition task we address has some overlap with related NLP al- gorithms, but each falls short in one or more re- spects. Elementary discourse unit (EDU) …

Content truncated to 500 words