Sentence Representations via Gaussian Embedding

arXiv:2305.12990v2 Announce Type: replace
Abstract: Recent progress in sentence embedding, which represents the meaning of a sentence as a point in a vector space, has achieved high performance on tasks such as a semantic textual similarity (STS) task. However, sentence representations as a point in a vector space can express only a part of the diverse information that sentences have, such as asymmetrical relationships between sentences. This paper proposes GaussCSE, a Gaussian distribution-based contrastive learning framework for sentence embedding that …

abstract arxiv cs.cl embedding express meaning part performance progress semantic space tasks textual type vector via