@inproceedings{brazinskas-etal-2018-embedding,
title = "Embedding Words as Distributions with a {B}ayesian Skip-gram Model",
author = "Bra{\v{z}}inskas, Arthur and
Havrylov, Serhii and
Titov, Ivan",
booktitle = "Proceedings of the 27th International Conference on Computational Linguistics",
month = aug,
year = "2018",
address = "Santa Fe, New Mexico, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/C18-1151",
pages = "1775--1789",
abstract = "We introduce a method for embedding words as probability densities in a low-dimensional space. Rather than assuming that a word embedding is fixed across the entire text collection, as in standard word embedding methods, in our Bayesian model we generate it from a word-specific prior density for each occurrence of a given word. Intuitively, for each word, the prior density encodes the distribution of its potential {`}meanings{'}. These prior densities are conceptually similar to Gaussian embeddings of {\.e}wcitevilnis2014word. Interestingly, unlike the Gaussian embeddings, we can also obtain context-specific densities: they encode uncertainty about the sense of a word given its context and correspond to the approximate posterior distributions within our model. The context-dependent densities have many potential applications: for example, we show that they can be directly used in the lexical substitution task. We describe an effective estimation method based on the variational autoencoding framework. We demonstrate the effectiveness of our embedding technique on a range of standard benchmarks.",
}