Controllable Lexical Simplification for English
This work addresses lexical simplification, a specific NLP task, by applying a fine-tuning approach previously unused in this area, representing an incremental advancement.
The authors tackled lexical simplification for English by fine-tuning T5, achieving performance comparable to the state-of-the-art LSBert and outperforming it in some cases on datasets like LexMTurk, BenchLS, and NNSeval.
Fine-tuning Transformer-based approaches have recently shown exciting results on sentence simplification task. However, so far, no research has applied similar approaches to the Lexical Simplification (LS) task. In this paper, we present ConLS, a Controllable Lexical Simplification system fine-tuned with T5 (a Transformer-based model pre-trained with a BERT-style approach and several other tasks). The evaluation results on three datasets (LexMTurk, BenchLS, and NNSeval) have shown that our model performs comparable to LSBert (the current state-of-the-art) and even outperforms it in some cases. We also conducted a detailed comparison on the effectiveness of control tokens to give a clear view of how each token contributes to the model.