We introduce a new type of 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). Our word vectors are learned functions of the internal states of a deep bidirectional language model (biLM), which is pre-trained on a large text corpus. We show that these representations can be easily added to existing models and significantly improve the state of the art across six challenging NLP problems, including question answering, textual entailment and sentiment analysis. We also present an analysis showing that exposing the deep internals of the pre-trained network is crucial, allowing downstream models to mix different types of semi-supervision signals.
核心贡献 · Key contributions
提出 ELMo，一种基于双向语言模型的深度上下文词表示。 Introduces ELMo, deep contextualized word representations from a bidirectional language model.
展示 ELMo 在六个 NLP 任务（问答、文本蕴含、语义角色标注、共指消解、命名实体识别、情感分析）上提升最先进水平。 Shows ELMo improves state-of-the-art on six NLP tasks: QA, textual entailment, SRL, coreference, NER, sentiment.
证明使用所有 biLM 层优于仅使用顶层。 Demonstrates that using all biLM layers outperforms using only the top layer.