šThe Current Best of Universal Word Embeddings and Sentence Embeddings
A Chinese version of this article can be found here, thanks to Jakukyo.
Word and sentence embeddings have become an essential part of any Deep-Learning-based natural language processing systems.
They encode words and sentences š in fixed-length dense vectors š to drastically improve the processing of textual data.
A huge trend is the quest for Universal Embeddings: embeddings that are pre-trained on a large corpus and can be plugged in a variety of downstream task models (sentimental analysis, classification, translationā¦) to automatically improve their performance by incorporating some general word/sentence representations learned on the larger dataset.
Itās a form of transfer learning. Transfer learning has been recently shown to drastically increase the performance of NLP models on important tasks such as text classification. Go check the very nice work of Jeremy Howard and Sebastian Ruder (ULMFiT) to see it in action.
While unsupervised representation learning of sentences had been the norm for quite some time, the last few months have seen a shift toward supervised and multi-task learning schemes with a number of very interesting proposals in late 2017/early 2018.
This post is thus a brief primer on the current state-of-the-art in Universal Word and Sentence Embeddings, detailing a few
- strong/fast baselines: FastText, Bag-of-Words
- state-of-the-art models: ELMo, Skip-Thoughts, Quick-Thoughts, InferSent, MILA/MSRās General Purpose Sentence Representations & Googleās Universal Sentence Encoder.
If you want some background on what happened before 2017 š, I recommend the nice post on word embeddings that Sebastian wrote last year and his intro posts.
Letās start with word embeddings.
Recent Developments in Word Embeddings
A wealth of possible ways to embed words have been proposed over the last five years. The most commonly used models are word2vec and GloVe which are both unsupervised approaches based on the distributional hypothesis (words that occur in the same contexts tend to have similar meanings).
While several works augment these unsupervised approaches by incorporating the supervision of semantic or syntactic knowledge, purely unsupervised approaches have seen interesting developments in 2017ā2018, the most notable being FastText (an extension of word2vec) and ELMo (state-of-the-art contextual word vectors).
FastText was developed by the team of Tomas Mikolov who proposed the word2vec framework in 2013, triggering the explosion of research on universal word embeddings.
The main improvement of FastText over the original word2vec vectors is the inclusion of character n-grams, which allows computing word representations for words that did not appear in the training data (āout-of-vocabularyā words).
FastText vectors are super-fast to train and are available in 157 languages trained on Wikipedia and Crawl. They are a great baseline.
The Deep Contextualized Word Representations (ELMo) have recently improved the state of the art in word embeddings by a noticeable amount. They were developed by the Allen institute for AI and will be presented at NAACL 2018 in early June.
In ELMo, each word is assigned a representation which is a function of the entire corpus sentences to which they belong. The embeddings are computed from the internal states of a two-layers bidirectional Language Model (LM), hence the name āELMoā: Embeddings from Language Models.
Specificities of ELMo:
- ELMoās inputs are characters rather than words. They can thus take advantage of sub-word units to compute meaningful representations even for out-of-vocabulary words (like FastText).
- ELMo are concatenations of the activations on several layers of the biLMs. Different layers of a language model encode different kind of information on a word (e.g. Part-Of-Speech tagging is well predicted by the lower level layers of a biLSTM while word-sense disambiguation is better encoded in higher-levels). Concatenating all layers allows to freely combine a variety of word representations for better performances on downstream tasks.
Now, letās turn to universal sentence embeddings.
The Rise of Universal Sentence Embeddings
There are currently many competing schemes for learning sentence embeddings. While simple baselines like averaging word embeddings consistently give strong results, a few novel unsupervised and supervised approaches, as well as multi-task learning schemes, have emerged in late 2017-early 2018 and lead to interesting improvements.
Letās go quickly through the four types of approaches currently studied: from simple word vector averaging baselines to unsupervised/supervised approaches and multi-task learning schemes (as illustrated above).
There is a general consensus in the field that the simple approach of directly averaging a sentenceās word vectors (so-called Bag-of-Word approach) gives a strong baseline for many downstream tasks.
A good algorithm for computing such a baseline is detailed in the work of Arora et al. published last year at ICLR, A Simple but Tough-to-Beat Baseline for Sentence Embeddings: use a popular word embeddings of your choice, encode a sentence in a linear weighted combination the word vectors and perform a common component removal (remove the projection of the vectors on their first principal component). This general method has deeper and powerful theoretical motivations that rely on a generative model which uses a random walk on a discourse vector to generate text (we wonāt discuss the theoretical details here).
A very recent implementation of a strong Bag-of-Word baseline (even stronger than Aroraās one) is the Concatenated p-mean Embeddings from the University of Darmstadt that you will find here (thanks Yaser for pointing that work out).
Going beyond simple averaging, the first major proposals were using unsupervised training objectives, starting with the Skip-thoughts vectors proposed by Jamie Kiros and co-workers in 2015.
Unsupervised schemes learn sentence embeddings as a byproduct of learning to predict a coherent succession of sentences or a coherent succession of clauses inside a sentence. These approaches can (in theory) make use of any text dataset as long as it contains sentences/clauses juxtaposed in a coherent way.
Skip-thoughts vectors is the archetypical example of learning unsupervised sentence embeddings. It can be though as the equivalent for sentences of the skip-gram model developed for word embeddings: rather than predicting the words surrounding a word, we try to predict the surroundings sentences of a given sentence. The model consists in an RNN-based encoder-decoder which is trained to reconstruct the surrounding sentences from the current sentence.
One interesting insight in the Skip-Thought paper was a vocabulary expansion scheme: Kiros et al. handled words not seen during training by learning a linear transformation between their RNN word embedding space and a larger word embedding such as word2vec.
Quick-thoughts vectors are a recent development of the Skip-thoughts vectors, presented this year at ICLR. In this work, the task of predicting the next sentence given the previous one is reformulated as a classification task: the decoder is replaced by a classifier which has to choose the next sentence among a set of candidates. It can be interpreted as a discriminative approximation to the generation problem.
One strength of this model is its speed of training (an order of magnitude compared to Skip-thoughts model) making it a competitive solution to exploit massive dataset.
For a long time, supervised learning of sentence embeddings was thought to give lower-quality embeddings than unsupervised approaches but this assumption has recently been overturned, in part following the publication of the InferSent results.
Unlike the unsupervised approaches detailed before, supervised learning requires a labelled dataset annotated for some task like Natural Language Inference (e.g. with pairs of entailed sentences) or Machine Translation (with pairs of translated sentences) which poses the question of the specific task to choose and the related question of the size of the dataset required for good quality embeddings. We talk more about these questions in the next and last section on Multi-task learning but before that, letās see whatās behind the InferSent breakthrough that was published in 2017.
InferSent is an interesting approach by the simplicity of its architecture. It uses the Stanford Natural Language Inference (SNLI) Corpus (a set of of 570k pairs of sentences labelled with 3 categories: neutral, contradiction and entailment) to train a classifier on top of a sentence encoder. Both sentences are encoded using the same encoder while the classifier is trained on a pair representation constructed from the two sentence embeddings. Conneau et al. adopt a bi-directional LSTM completed with a max-pooling operator as sentence encoder.
The success of InferSent lead poses the following question in addition to the usual quest for selecting the best neural net model:
Which supervised training task would learn sentence embeddings that better generalize on downstream tasks?
Multi-task learning can be seen as a generalization of Skip-Thoughts, InferSent, and the related unsupervised/supervised learning schemes, that answer this question by trying to combine several training objectives in one training scheme.
Several recent proposals for multi-task learning were published in early 2018. Letās quickly go through MILA/MSRās General Purpose Sentence Representation and Googleās Universal Sentence Encoder.
In the paper describing MILA & Microsoft Montrealās work and presented at ICLR 2018 (Learning General Purpose Distributed Sentence Representation via Large Scale Multi-Task Learning), Subramanian et al observe that to be able to generalize over a wide range of diverse tasks, it is necessary to encode multiple aspects of the same sentence.
The authors thus leverage a one-to-many multi-tasking learning framework to learn a universal sentence embedding by switching between several tasks. The 6 tasks chosen (Skip-thoughts prediction of the next/previous sentence, neural machine translation, constituency parsing and natural language inference) share the same sentence embedding obtained by a bi-directional GRU. Experiments suggest that syntactic properties are better learned when adding a multi-language neural machine translation task, length and word order are learned with a parsing task and training a natural language inference encodes syntax information.
Googleās Universal Sentence Encoder, published in early 2018, follows the same approach. Their encoder uses a transformer-network that is trained on a variety of data sources and a variety of tasks with the aim of dynamically accommodating a wide variety of natural language understanding tasks. A pre-trained version has been made available for TensorFlow.
This concludes our short summary on the current state of Universal Words and Sentence Embeddings.
The domain has seen a lot of interesting developments in the last few months together with great progresses in the ways we assess and probe the performance of these embeddings and their inherent bias/fairness (a real issue when you talk about Universal Embeddings). We didnāt have time to talk about these latest topics but you can find a few links in the references.
I hope you enjoyed this brief!
Clap š a couple of times if you liked it and want us to post more of these!
Some references
- Very recently, C. Perone and co-workers published a nice and extensive comparison between ELMo, InferSent, Google Universal Sentence Encoder, p-mean, Skip-thought, etc. Here is a link to the paper: https://arxiv.org/abs/1806.06259
- A nice ressource on traditional word embeddings like word2vec, GloVe and their supervised learning augmentations is the github repository of Hironsan. More recent developments are FastText and ELMo.
- Sentence embeddings papers: Skip-Thoughts, Quick-Thoughts, DiscSent, InferSent, MILA/MSRās General Purpose Sentence Representations, Googleās Universal Sentence Encoder & Google Input-Ouput Sentence learning on dialog.
- If youāre interested in the way we evaluate sentence embeddings, you should definitely check the recent work of Facebook on SentEval and its probing tasks as well as the recently published GLUE benchmark by NYU, UW and DeepMind researchers.