Abstract

We present models for encoding sentences into embedding vectors that specifically target transfer learning to other NLP tasks. The models are efficient and result in accurate performance on diverse transfer tasks. Two variants of the encoding models allow for trade-offs between accuracy and compute resources. For both variants, we investigate and report the relationship between model complexity, resource consumption, the availability of transfer task training data, and task performance. Comparisons are made with baselines that use word level transfer learning via pretrained word embeddings as well as baselines do not use any transfer learning. We find that transfer learning using sentence embeddings tends to outperform word level transfer. With transfer learning via sentence embeddings, we observe surprisingly good performance with minimal amounts of supervised training data for a transfer task. We obtain encouraging results on Word Embedding Association Tests (WEAT) targeted at detecting model bias. Our pre-trained sentence encoding models are made freely available for download and on TF Hub.

Keywords

Computer scienceTransfer of learningSentenceEncoderEncoding (memory)Word (group theory)Task (project management)Natural language processingEmbeddingArtificial intelligenceTransfer (computing)Speech recognitionMachine learningMathematics

Related Publications

Publication Info

Year
2018
Type
preprint
Citations
1289
Access
Closed

External Links

View on DOI.org

Social Impact

Altmetric
PlumX Metrics

Social media, news, blog, policy document mentions

Citation Metrics

1289
OpenAlex

Cite This

Daniel Cer, Yinfei Yang, Sheng-yi Kong et al. (2018). Universal Sentence Encoder. arXiv (Cornell University) . https://doi.org/10.48550/arxiv.1803.11175

Identifiers

DOI
10.48550/arxiv.1803.11175