Embeddings from Language Models
Year: 2,018
Journal: International Conference on Learning Representations Allen Institute for Artificial Intelligence
Languages: Basque, English, German, Japanese, Portuguese
Programming languages: Python
Input data:
character-based
Output data:
word representations as functions of the entire input sentence
Project website: https://allennlp.org/elmohttps://github.com/allenai/allennlp
ELMo is a deep contextualized word representation that models both (1) complex characteristics of word use (e.g., syntax and semantics), and (2) how these uses vary across linguistic contexts (i.e., to model polysemy).