Improving sentence compression by learning to predict gaze

Proceedings of NAACL-HLT 2016, pages 1528–1533, San Diego, California, June 12-17, 2016. c⃝2016 Association for Computational Linguistics Improving sentence compression by learning to predict gaze Sigrid Klerke University of Copenhagen skl@hum.ku.dk Yoav Goldberg Bar-Ilan University yoav.goldberg@gmail.com Anders Søgaard University of Copenhagen soegaard@hum.ku.dk Abstract We show how eye-tracking corpora can be used to improve sentence compression mod- els, presenting a novel multi-task learning al- gorithm based on multi-layer LSTMs. We ob- tain performance competitive with or better than state-of-the-art approaches. 1 Introduction Sentence compression is a basic operation in text simplification which has the potential to improve statistical machine translation and automatic sum- marization (Berg-Kirkpatrick et al., 2011; Klerke et al., 2015), as well as helping poor readers in need of assistive technologies (Canning et al., 2000). This work suggests using eye-tracking recordings for im- proving sentence compression for text simplification systems and is motivated by two observations: (i) Sentence compression is the task of automatically making sentences easier to process by shortening them. (ii) Eye-tracking measures such as first-pass reading time and time spent on regressions, i.e., dur- ing second and later passes over the text, are known to correlate with perceived text difficulty (Rayner et al., 2012). These two observations recently lead Klerke et al. (2015) to suggest using eye-tracking measures as metrics in text simplification. We go beyond this by suggesting that eye-tracking recordings can be used to induce better models for sentence compression for text simplification. Specifically, we show how to use existing eye-tracking recordings to improve the induction of Long Short-Term Memory models (LSTMs) for sentence compression. Our proposed model does not require that the gaze data and the compression data come from the same source. Indeed, in this work we use gaze data from readers of the Dundee Corpus to improve sentence compression results on several datasets. While not explored here, an intriguing potential of this work is in deriving sentence simplification models that are personalized for individual users, based on their reading behavior. Several approaches to sentence compression have been proposed, from noisy channel models (Knight and Marcu, 2002) over conditional random fields (Elming et al., 2013) to tree-to-tree machine trans- lation models (Woodsend and Lapata, 2011). More recently, Filippova et al. (2015) successfully used LSTMs for sentence compression on a large scale parallel dataset. We do not review the literature here, and only compare to Filippova et al. (2015). Our contributions • We present a novel multi-task learning ap- proach to sentence compression using labelled data for sentence compression and a disjoint eye-tracking corpus. • Our method is fully competitive with state-of- the-art across three corpora. • Our code is made publicly available https://bitbucket.org/soegaard/at gaze-mtl16. 2 Gaze during reading Readers fixate longer at rare words, words that are semantically ambiguous, and words that are mor- 1528 phologically complex (Rayner et al., 2012). These are also words that are likely to be replaced with simpler ones in sentence simplification, but it is not clear that they are words that would necessarily be removed in …

Content truncated to 500 words