AutoExtend: Extending Word Embeddings to Embeddings for Synsets and\n Lexemes

Abstract

We present \\textit{AutoExtend}, a system to learn embeddings for synsets and\nlexemes. It is flexible in that it can take any word embeddings as input and\ndoes not need an additional training corpus. The synset/lexeme embeddings\nobtained live in the same vector space as the word embeddings. A sparse tensor\nformalization guarantees efficiency and parallelizability. We use WordNet as a\nlexical resource, but AutoExtend can be easily applied to other resources like\nFreebase. AutoExtend achieves state-of-the-art performance on word similarity\nand word sense disambiguation tasks.\n