Learning Universal Sentence Representations with Mean-Max Attention Autoencoder

TL;DR

This paper introduces a mean-max attention autoencoder that uses self-attention mechanisms to learn universal sentence representations efficiently, outperforming previous models on various transfer tasks.

Contribution

It presents a novel mean-max attention autoencoder leveraging self-attention for unsupervised sentence embedding, reducing training time and improving performance.

Findings

Outperforms state-of-the-art unsupervised methods on 10 transfer tasks.

Reduces training time compared to RNN-based models.

Effective in capturing diverse sentence information.

Abstract

In order to learn universal sentence representations, previous methods focus on complex recurrent neural networks or supervised learning. In this paper, we propose a mean-max attention autoencoder (mean-max AAE) within the encoder-decoder framework. Our autoencoder rely entirely on the MultiHead self-attention mechanism to reconstruct the input sequence. In the encoding we propose a mean-max strategy that applies both mean and max pooling operations over the hidden vectors to capture diverse information of the input. To enable the information to steer the reconstruction process dynamically, the decoder performs attention over the mean-max representation. By training our model on a large collection of unlabelled data, we obtain high-quality representations of sentences. Experimental results on a broad range of 10 transfer tasks demonstrate that our model outperforms the state-of-the-art unsupervised single methods, including the classical skip-thoughts and the advanced skip-thoughts+LN model. Furthermore, compared with the traditional recurrent neural network, our mean-max AAE greatly reduce the training time.