Incorporating Word Sense Disambiguation in Neural Language Models
This work addresses the challenge of enhancing language understanding for NLP applications, though it is incremental as it builds on existing models with minor modifications.
The authors tackled the problem of improving neural language models by incorporating gloss definitions from lexical resources, resulting in a 0.5% F1 increase on the SemCor 3.0 dataset for Word Sense Disambiguation and a 1.1% average boost on the GLUE benchmark for BERT.
We present two supervised (pre-)training methods to incorporate gloss definitions from lexical resources into neural language models (LMs). The training improves our models' performance for Word Sense Disambiguation (WSD) but also benefits general language understanding tasks while adding almost no parameters. We evaluate our techniques with seven different neural LMs and find that XLNet is more suitable for WSD than BERT. Our best-performing methods exceeds state-of-the-art WSD techniques on the SemCor 3.0 dataset by 0.5% F1 and increase BERT's performance on the GLUE benchmark by 1.1% on average.