Neural CRF Model for Sentence Alignment in Text Simplification
This work addresses the need for high-quality training data in text simplification systems, which is crucial for researchers and developers in natural language processing, though it is incremental as it builds on existing alignment methods.
The authors tackled the problem of sentence alignment for text simplification by creating two manually annotated datasets and proposing a neural CRF model, which outperformed previous work by over 5 points in F1 and led to new datasets that improved state-of-the-art simplification results.
The success of a text simplification system heavily depends on the quality and quantity of complex-simple sentence pairs in the training corpus, which are extracted by aligning sentences between parallel articles. To evaluate and improve sentence alignment quality, we create two manually annotated sentence-aligned datasets from two commonly used text simplification corpora, Newsela and Wikipedia. We propose a novel neural CRF alignment model which not only leverages the sequential nature of sentences in parallel documents but also utilizes a neural sentence pair model to capture semantic similarity. Experiments demonstrate that our proposed approach outperforms all the previous work on monolingual sentence alignment task by more than 5 points in F1. We apply our CRF aligner to construct two new text simplification datasets, Newsela-Auto and Wiki-Auto, which are much larger and of better quality compared to the existing datasets. A Transformer-based seq2seq model trained on our datasets establishes a new state-of-the-art for text simplification in both automatic and human evaluation.