English

Not All Neural Embeddings are Born Equal

Computation and Language 2014-11-14 v2

Abstract

Neural language models learn word representations that capture rich linguistic and conceptual information. Here we investigate the embeddings learned by neural machine translation models. We show that translation-based embeddings outperform those learned by cutting-edge monolingual models at single-language tasks requiring knowledge of conceptual similarity and/or syntactic role. The findings suggest that, while monolingual models learn information about how concepts are related, neural-translation models better capture their true ontological status.

Keywords

Cite

@article{arxiv.1410.0718,
  title  = {Not All Neural Embeddings are Born Equal},
  author = {Felix Hill and KyungHyun Cho and Sebastien Jean and Coline Devin and Yoshua Bengio},
  journal= {arXiv preprint arXiv:1410.0718},
  year   = {2014}
}

Comments

4 pages plus 1 page of references

R2 v1 2026-06-22T06:12:08.153Z