Unsupervised Learning of Sentence Embeddings using Compositional n-Gram Features

Matteo Pagliardini, Prakhar Gupta, Martin Jaggi
3/7/2017
cs.CLcs.AIcs.IR

Official code: epfml/sent2vec (1,202 stars, last commit 2022-08-03).

Abstract

The recent tremendous success of unsupervised word embeddings in a multitude of applications raises the obvious question if similar methods could be derived to improve embeddings (i.e. semantic representations) of word sequences as well. We present a simple but efficient unsupervised objective to train distributed representations of sentences. Our method outperforms the state-of-the-art unsupervised models on most benchmark tasks, highlighting the robustness of the produced general-purpose sentence embeddings.

View on arXivView PDF

Code Implementations(1)

epfml/sent2vecOfficial100%

General purpose unsupervised sentence representations

1,202257C++, CythonMar 15, 20174 years agoNOASSERTION

Similar Papers

Learning General Purpose Distributed Sentence Representations via Large Scale Multi-task Learning

86% match2 repos2.1k2018

Supervised Learning of Universal Sentence Representations from Natural Language Inference Data

85% match2 repos2.3k2017

A Simple and Plug-and-play Method for Unsupervised Sentence Representation Enhancement

83% match2023

TSDAE: Using Transformer-based Sequential Denoising Auto-Encoder for Unsupervised Sentence Embedding Learning

83% match2 repos712021

UNSEE: Unsupervised Non-contrastive Sentence Embeddings

83% match2024

Static Fuzzy Bag-of-Words: a lightweight sentence embedding algorithm

82% match2023

Cite this paper

@article{pagliardini2017unsupervised,
  title  = {Unsupervised Learning of Sentence Embeddings using Compositional n-Gram Features},
  author = {Matteo Pagliardini and Prakhar Gupta and Martin Jaggi},
  year   = {2017},
  eprint = {1703.02507},
  archivePrefix = {arXiv},
  url    = {https://arxiv.org/abs/1703.02507}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/1703.02507.svg)](https://codeofpaper.com/paper/1703.02507)

Discussion