Sentence Meta-Embeddings for Unsupervised Semantic Textual Similarity
Computation and Language
2020-06-25 v3
Abstract
We address the task of unsupervised Semantic Textual Similarity (STS) by ensembling diverse pre-trained sentence encoders into sentence meta-embeddings. We apply, extend and evaluate different meta-embedding methods from the word embedding literature at the sentence level, including dimensionality reduction (Yin and Sch\"utze, 2016), generalized Canonical Correlation Analysis (Rastogi et al., 2015) and cross-view auto-encoders (Bollegala and Bao, 2018). Our sentence meta-embeddings set a new unsupervised State of The Art (SoTA) on the STS Benchmark and on the STS12-STS16 datasets, with gains of between 3.7% and 6.4% Pearson's r over single-source systems.
Cite
@article{arxiv.1911.03700,
title = {Sentence Meta-Embeddings for Unsupervised Semantic Textual Similarity},
author = {Nina Poerner and Ulli Waltinger and Hinrich Schütze},
journal= {arXiv preprint arXiv:1911.03700},
year = {2020}
}