Improving Human Text Comprehension through Semi-Markov CRF-based Neural Section Title Generation

Proceedings of NAACL-HLT 2019, pages 1677–1688 Minneapolis, Minnesota, June 2 - June 7, 2019. c⃝2019 Association for Computational Linguistics 1677 Improving Human Text Comprehension through Semi-Markov CRF-based Neural Section Title Generation Sebastian Gehrmann1,2, Steven Layne2,3, Franck Dernoncourt2 1Harvard SEAS, 2Adobe Research, 3University of Illinois Urbana Champaign gehrmann@seas.harvard.edu, solayne2@illinois.edu, franck.dernoncourt@adobe.com Abstract Titles of short sections within long documents support readers by guiding their focus towards relevant passages and by providing anchor- points that help to understand the progression of the document. The positive effects of sec- tion titles are even more pronounced when measured on readers with less developed read- ing abilities, for example in communities with limited labeled text resources. We, therefore, aim to develop techniques to generate sec- tion titles in low-resource environments. In particular, we present an extractive pipeline for section title generation by first selecting the most salient sentence and then applying deletion-based compression. Our compression approach is based on a Semi-Markov Con- ditional Random Field that leverages unsu- pervised word-representations such as ELMo or BERT, eliminating the need for a com- plex encoder-decoder architecture. The results show that this approach leads to competitive performance with sequence-to-sequence mod- els with high resources, while strongly outper- forming it with low resources. In a human- subjects study across subjects with varying reading abilities, we find that our section titles improve the speed of completing comprehen- sion tasks while retaining similar accuracy. 1 Introduction Section titles in long documents that explain the content of the section improve the recall of con- tent (Dooling and Lachman, 1971; Smith and Swinney, 1992) while simultaneously increasing the reading speed (Bransford and Johnson, 1972). Additionally, they can provide a context to al- low ambiguous words to be understood more eas- ily (Wiley and Rayner, 2000) and to better un- derstand the overall text (Kintsch and Van Dijk, 1978). However, most documents do not include titles for short segments or only provide a very ab- stract description of their topics, e.g. “Geography” When another old cave is discovered in the south of France, it is not usually news. Rather, it is an ordinary event. Such discoveries are so frequent these days that hardly anybody pays heed to them. However, when the Lascaux cave com- plex was discovered in 1940, the world was amazed. Painted directly on its walls were hundreds of scenes showing how people lived thousands of years ago. The scenes show people hunting animals, such as bison or wild cats. Other images depict birds and, most noticeably, horses, which appear in more than 300 wall images, by far outnumbering all other animals. Early artists drawing these animals accomplished a monumental and difficult task. They did not limit themselves to the easily accessible walls but carried their painting materials to spaces that required climbing steep walls or crawling into narrow passages in the Lascaux complex. Unfortunately, the paintings have been exposed to the destructive action of water and temperature changes, which easily wear the images away. Because the Lascaux caves have many entrances, air …

Content truncated to 500 words