natural-language-processing

Sentence Embedding

3260 papers • 126 benchmarks • 313 datasets

This task has no description! Would you like to contribute one?

(Image credit: Papersgraph)

Benchmarks

These leaderboards are used to track progress in sentence-embedding

Trend

Dataset

Best Model

Actions

No benchmarks available.

Libraries

i

Use these libraries to find sentence-embedding models and implementations

UKPLab/sentence-transformers

2 papers 13,689

Datasets

Subtasks

No subtasks available.

Most implemented papers

Evaluation of sentence embeddings in downstream and linguistic probing tasks

C. Perone, Roberto Silveira, Thomas S. Paula•Fri Jun 15 2018

It is shown that a simple approach using bag-of-words with a recently introduced language model for deep context-dependent word embeddings proved to yield better results in many tasks when compared to sentence encoders trained on entailment datasets.

160

Content

COSTRA 1.0

0

Paper Graph

A Structured Self-attentive Sentence Embedding

Yoshua Bengio, Zhouhan Lin, Bowen Zhou, C. D. Santos, Bing Xiang, Mo Yu, Minwei Feng•Wed Mar 08 2017

A new model for extracting an interpretable sentence embedding by introducing self-attention is proposed, which uses a 2-D matrix to represent the embedding, with each row of the matrix attending on a different part of the sentence.

2264 0

Paper Graph

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

Iryna Gurevych, Nils Reimers, Kexin Wang•Tue Apr 13 2021

A new state-of-the-art unsupervised method based on pre-trained Transformers and Sequential Denoising Auto-Encoder (TSDAE) which outperforms previous approaches by up to 6.4 points and can achieve up to 93.1% of the performance of in-domain supervised approaches.

221 0

Paper Graph

Language-agnostic BERT Sentence Embedding

Daniel Matthew Cer, Wei Wang, Yinfei Yang, Fangxiaoyu Feng, N. Arivazhagan•Thu Jul 02 2020

It is shown that introducing a pre-trained multilingual language model dramatically reduces the amount of parallel training data required to achieve good performance by 80%, and a model that achieves 83.7% bi-text retrieval accuracy over 112 languages on Tatoeba is released.

1153 0

Paper Graph

Aligning Books and Movies: Towards Story-Like Visual Explanations by Watching Movies and Reading Books

R. Zemel, A. Torralba, R. Urtasun, R. Salakhutdinov, S. Fidler, Ryan Kiros, Yukun Zhu•Sun Jun 21 2015

To align movies and books, a neural sentence embedding that is trained in an unsupervised way from a large corpus of books, as well as a video-text neural embedding for computing similarities between movie clips and sentences in the book are proposed.

2688 0

Paper Graph

ColBERT: Using BERT sentence embedding in parallel neural networks for computational humor

Issa Annamoradnejad, Gohar Zoghi•Sun Apr 26 2020

The evaluation performed on two contrasting settings confirm the strength and robustness of the model and suggests two important factors in achieving high accuracy in the current task: usage of sentence embeddings and utilizing the linguistic structure of humor in designing the proposed model.

36 0

Paper Graph

DiSAN: Directional Self-Attention Network for RNN/CNN-free Language Understanding

Shirui Pan, Tianyi Zhou, Tao Shen, Guodong Long, Jing Jiang, Chengqi Zhang•Wed Sep 13 2017

A novel attention mechanism in which the attention between elements from input sequence(s) is directional and multi-dimensional (i.e., feature-wise) and a light-weight neural net is proposed, based solely on the proposed attention without any RNN/CNN structure, which outperforms complicated RNN models on both prediction quality and time efficiency.

778 0

Paper Graph

Fine-grained Analysis of Sentence Embeddings Using Auxiliary Prediction Tasks

Yoav Goldberg, Yonatan Belinkov, Yossi Adi, Einat Kermany, Ofer Lavi•Sun Aug 14 2016

This work proposes a framework that facilitates better understanding of the encoded representations of sentence vectors and demonstrates the potential contribution of the approach by analyzing different sentence representation mechanisms.

568 0

Paper Graph

SBERT-WK: A Sentence Embedding Method by Dissecting BERT-Based Word Models

C.-C. Jay Kuo, Bin Wang•Sat Feb 15 2020

This work proposes a new sentence embedding method by dissecting BERT-based word models through geometric analysis of the space spanned by the word representation, called SBERT-WK, which achieves the state-of-the-art performance.

184 0

Paper Graph

Making Monolingual Sentence Embeddings Multilingual Using Knowledge Distillation

Iryna Gurevych, Nils Reimers•Mon Apr 20 2020

An easy and efficient method to extend existing sentence embedding models to new languages by using the original (monolingual) model to generate sentence embeddings for the source language and then training a new system on translated sentences to mimic the original model.

1218 0

Paper Graph

Adding a benchmark result helps the community track progress.