Word Sense Disambiguation with LSTM: Do We Really Need 100 Billion Words?

2017/12/09 by Le, Minh, Postma, Marten, Urbani, Jacopo
#Computation and Language (cs.CL) #FOS: Computer and information sciences

paper · doi:10.48550/arxiv.1712.03376

Abstract

Recently, Yuan et al. (2016) have shown the effectiveness of using Long Short-Term Memory (LSTM) for performing Word Sense Disambiguation (WSD). Their proposed technique outperformed the previous state-of-the-art with several benchmarks, but neither the training data nor the source code was released. This paper presents the results of a reproduction study of this technique using only openly available datasets (GigaWord, SemCore, OMSTI) and software (TensorFlow). From them, it emerged that state-of-the-art results can be obtained with much less data than hinted by Yuan et al. All code and trained models are made freely available.

Related

• Word Sense Disambiguation using a Bidirectional LSTM
• One Single Deep Bidirectional LSTM Network for Word Sense Disambiguation\n of Text Data
• Exploring the Word Sense Disambiguation Capabilities of Large Language Models
• Using Two Losses and Two Datasets Simultaneously to Improve TempoWiC Accuracy
• A Comparison between Supervised Learning Algorithms for Word Sense Disambiguation