Unsupervised Learning of Sentence Embeddings using Compositional n-Gram Features
This addresses the need for robust general-purpose sentence embeddings in natural language processing, though it appears incremental as it builds on existing word embedding methods.
The paper tackles the problem of deriving unsupervised sentence embeddings, presenting a simple objective that outperforms state-of-the-art unsupervised models on most benchmark tasks.
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.