natural-language-processing-2

Word Similarity

3260 papers • 126 benchmarks • 313 datasets

Calculate a numerical score for the semantic similarity between two words.

(Image credit: Papersgraph)

Benchmarks

These leaderboards are used to track progress in word-similarity-2

Trend

Dataset

Best Model

Actions

No benchmarks available.

Libraries

i

Use these libraries to find word-similarity-2 models and implementations

dmlc/gluon-nlp

2 papers 2,548

Datasets

AnlamVer

Bangla Word Analogy

Subtasks

No subtasks available.

Most implemented papers

Enriching Word Vectors with Subword Information

Edouard Grave, Piotr Bojanowski, Armand Joulin, Tomas Mikolov•Thu Jul 14 2016

A new approach based on the skipgram model, where each word is represented as a bag of character n-grams, with words being represented as the sum of these representations, which achieves state-of-the-art performance on word similarity and analogy tasks.

10521

Content

LuminosoInsight/conceptnet-vector-e…

2 papers 1,267

0

Paper Graph

All-but-the-Top: Simple and Effective Postprocessing for Word Representations

P. Viswanath, Jiaqi Mu, S. Bhat•Sat Feb 04 2017

This paper demonstrates a counter-intuitive, postprocessing technique -- eliminate the common mean vector and a few top dominating directions from the word vectors -- that renders off-the-shelf representations even stronger.

347 0

Paper Graph

How to evaluate word embeddings? On importance of data efficiency and simple supervised tasks

Stanislaw Jastrzebski, Wojciech M. Czarnecki, Damian Lesniak•Mon Feb 06 2017

It is proposed that evaluation of word representation evaluation should focus on data efficiency and simple supervised tasks, where the amount of available data is varied and scores of a supervised model are reported for each subset (as commonly done in transfer learning).

79 0

Paper Graph

Calculating the similarity between words and sentences using a lexical database and corpus statistics

Vijay K. Mago, Atish Pawar•Wed Feb 14 2018

The proposed method follows an edge-based approach using a lexical database and gives highest correlation value for both word and sentence similarity outperforming other similar models.

66 0

Paper Graph

Speech2Vec: A Sequence-to-Sequence Framework for Learning Word Embeddings from Speech

James R. Glass, Yu-An Chung•Thu Mar 22 2018

The proposed Speech2Vec model, a novel deep neural network architecture for learning fixed-length vector representations of audio segments excised from a speech corpus, is based on a RNN Encoder-Decoder framework, and borrows the methodology of skipgrams or continuous bag-of-words for training.

187 0

Paper Graph

Unsupervised Multilingual Word Embeddings

Claire Cardie, Xilun Chen•Tue Jul 31 2018

This work proposes a fully unsupervised framework for learning MWEs that directly exploits the relations between all language pairs and substantially outperforms previous approaches in the experiments on multilingual word translation and cross-lingual word similarity.

138 0

Paper Graph

SemGloVe: Semantic Co-Occurrences for GloVe From BERT

Yue Zhang, Fei Wu, Zhiyang Teng, Linchao Zhu, Leilei Gan, Yi Yang•Tue Dec 29 2020

SemGloVe is proposed, which distills semantic co-occurrences from BERT into static GloVe word embeddings and can define the co- Occurrence weights by directly considering the semantic distance between word pairs.

21 0

Paper Graph

WordRank: Learning Word Embeddings via Robust Ranking

Hyokun Yun, Pinar Yanardag, S. Vishwanathan, Shihao Ji, Shin Matsushima•Mon Jun 08 2015

This paper argues that word embedding can be naturally viewed as a ranking problem due to the ranking nature of the evaluation metrics, and proposes a novel framework WordRank that efficiently estimates word representations via robust ranking, in which the attention mechanism and robustness to noise are readily achieved via the DCG-like ranking losses.

39 0

Paper Graph

Definition Modeling: Learning to Define Word Embeddings in Natural Language

Chen Liang, Doug Downey, Thanapon Noraset, L. Birnbaum•Wed Nov 30 2016

The results show that a model that controls dependencies between the word being defined and the definition words performs significantly better, and that a character-level convolution layer that leverages morphology can complement word-level embeddings.

138 0

Paper Graph

Adding a benchmark result helps the community track progress.