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.
By Matteo Pagliardini, Prakhar Gupta, Martin Jaggi
Codes and Models are available at:
https://github.com/epfml/sent2vec
https://arxiv.org/abs/1703.02507
More from our feed
Key takeaways from the 2024 AI-Assisted Invention summit
Read more![Key takeaways from the 2024 AI Assisted Invention summit](https://www.iprova.com/wp-content/uploads/2024/06/key-takeaways-from-the-2024-AI-Assisted-Invention-Summit.jpg)
Iprova and Microsoft partner to launch AI-Assisted Invention Summit
Read more![](https://www.iprova.com/wp-content/uploads/2024/02/microsoft-iprova.png)
Lessons from MWC 2024
Read more![](https://www.iprova.com/wp-content/uploads/2024/03/1709024573330-834x556.jpg)