I made an introductory talk on word embeddings in the past and this write-up is an extended version of the part about philosophical ideas behind word vectors.
In natural language understanding, there is a hierarchy of lenses through which we can extract meaning - from words to sentences to paragraphs to documents. At the document level, one of the most useful ways to understand text is by analyzing its topics.
ConceptNet Numberbatch consists of state-of-the-art semantic vectors (also known as word embeddings) that can be used directly as a representation of word meanings or as a starting point for further machine learning.
M. Artetxe, G. Labaka, I. Lopez-Gazpio, and E. Agirre. Proceedings of the 22nd Conference on Computational Natural Language Learning, page 282--291. Association for Computational Linguistics, (2018)
S. Cordeiro, C. Ramisch, M. Idiart, and A. Villavicencio. Proceedings of the 54th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), 1, page 1986--1997. The Association for Computer Linguistics, (2016)
D. Tang, F. Wei, N. Yang, M. Zhou, T. Liu, and B. Qin. Proceedings of the 52nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), page 1555--1565. Baltimore, Maryland, Association for Computational Linguistics, (June 2014)