Jeffrey Pennington, Richard Socher, & Christopher Manning (2014)
Proceedings of the 2014 Conference on Empirical Methods in Natural Language Processing (EMNLP), 1532-1543.
DOI: https://doi.org/10.3115/v1/d14-1162
Abstract. Introduces GloVe, which learns word embeddings by factorising a global word-word co-occurrence matrix. The resulting vectors combine the statistical benefits of global matrix factorisation with the linear substructure of Word2Vec.
Tags: embeddings glove nlp