Abstract: Sentence splitting is a major simpliﬁcation operator. Here we present a simple and efﬁcient splitting algorithm based on an automatic semantic parser. After splitting, the text is amenable for further ﬁne-tuned simpliﬁcation operations. In particular, we show that neural Machine Translation can be effectively used in this situation. Previous application of Machine Translation for simpliﬁcation suffers from a considerable disadvantage in that they are over-conservative, often failing to modify the source in any way. Splitting based on semantic parsing, as proposed here, alleviates this issue. Extensive automatic and human evaluation shows that the proposed method compares favorably to the state-of-the-art in combined lexical and structural simpliﬁcation.
Authors: Elior Sulem, Omri Abend, Ari Rappoport (The Hebrew University of Jerusalem)