Table of Contents
Fetching ...

Edit-Constrained Decoding for Sentence Simplification

Tatsuya Zetsu, Yuki Arase, Tomoyuki Kajiwara

TL;DR

The proposed edit operation based lexically constrained decoding for sentence simplification by designing constraints that replicate the edit operations conducted in simplification and defining stricter satisfaction conditions consistently outperforms the previous studies.

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.

Edit-Constrained Decoding for Sentence Simplification

TL;DR

The proposed edit operation based lexically constrained decoding for sentence simplification by designing constraints that replicate the edit operations conducted in simplification and defining stricter satisfaction conditions consistently outperforms the previous studies.

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.
Paper Structure (29 sections, 5 equations, 2 figures, 7 tables)

This paper contains 29 sections, 5 equations, 2 figures, 7 tables.

Figures (2)

  • Figure 1: Our method constrains generation based on edit operations during beam search. Here, 'artisans' is replaced by 'craftsmen' by a substitution constraint.
  • Figure 2: Edit-based constraint decoding example (beam size is $3$)