Inducing and Embedding Senses with Scaled Gumbel Softmax

Abstract

Methods for learning word sense embeddings represent a single word with multiple sense-specific vectors. These methods should not only produce interpretable sense embeddings, but should also learn how to select which sense to use in a given context. We propose an unsupervised model that learns sense embeddings using a modified Gumbel softmax function, which allows for differentiable discrete sense selection. Our model produces sense embeddings that are competitive (and sometimes state of the art) on multiple similarity based downstream evaluations. However, performance on these downstream evaluations tasks does not correlate with interpretability of sense embeddings, as we discover through an interpretability comparison with competing multi-sense embeddings. While many previous approaches perform well on downstream evaluations, they do not produce interpretable embeddings and learn duplicated sense groups; our method achieves the best of both worlds.

References in corpus (9)

• Efficient Estimation of Word Representations in Vector Space
• Distributed Representations of Words and Phrases and their Compositionality
• Categorical Reparameterization with Gumbel-Softmax
• Towards A Rigorous Science of Interpretable Machine Learning
• word2vec Explained: deriving Mikolov et al.'s negative-sampling word-embedding method
• Enhanced LSTM for Natural Language Inference
• The Concrete Distribution: A Continuous Relaxation of Discrete Random Variables
• Breaking Sticks and Ambiguities with Adaptive Skip-gram
• Evaluation of sentence embeddings in downstream and linguistic probing tasks