Paraphrases do not explain word analogies
Not on the map. Why?
Abstract (arXiv)
Many types of distributional word embeddings (weakly) encode linguistic regularities as directions (the difference between "jump" and "jumped" will be in a similar direction to that of "walk" and "walked," and so on). Several attempts have been made to explain this fact. We respond to Allen and Hospedales' recent (ICML, 2019) theoretical explanation, which claims that word2vec and GloVe will encode linguistic regularities whenever a specific relation of paraphrase holds between the four words involved in the regularity. We demonstrate that the explanation does not go through: the paraphrase relations needed under this explanation do not hold empirically.
Key facts. Cited by 2 papers in the map (data through 7 Oct 2026). Builds on 6 of its 15 references that are in the map. First posted on arXiv 23 Feb 2021.
Builds on
6 of 15 references.
References · 6
- 2013Efficient Estimation of Word Representations in Vector Spacecited by 4,346
- 2019Analogies Explained: Towards Understanding Word Embeddingscited by 63
- 2018Towards Understanding Linear Word Analogiescited by 57
- 2015Take and Took, Gaggle and Goose, Book and Read: Evaluating the Utility of Vector Differences for Lexical Relation Learningcited by 29
- 2016Issues in evaluating semantic spaces using word analogiescited by 26
- 2020Analogies minus analogy test: measuring regularities in word embeddingscited by 17
Built on by
Citing papers per year: 2023: 2.
2 papers.
- 2023Paraphrase Types for Generation and Detection23 Oct 2023
- 2023Contrastive Loss is All You Need to Recover Analogies as Parallel Lines14 Jun 2023