Abstract
We consider the problem of learning general-purpose, paraphrastic sentence embeddings, revisiting the setting of Wieting et al. (2016b). While they found LSTM recurrent networks to underperform word averaging, we present several developments that together produce the opposite conclusion. These include training on sentence pairs rather than phrase pairs, averaging states to represent sequences, and regularizing aggressively. These improve LSTMs in both transfer learning and supervised settings. We also introduce a new recurrent architecture, the Gated Recurrent Averaging Network, that is inspired by averaging and LSTMs while outperforming them both. We analyze our learned models, finding evidence of preferences for particular parts of speech and dependency relations.- Anthology ID:
- P17-1190
- Volume:
- Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)
- Month:
- July
- Year:
- 2017
- Address:
- Vancouver, Canada
- Editors:
- Regina Barzilay, Min-Yen Kan
- Venue:
- ACL
- SIG:
- Publisher:
- Association for Computational Linguistics
- Note:
- Pages:
- 2078–2088
- Language:
- URL:
- https://aclanthology.org/P17-1190
- DOI:
- 10.18653/v1/P17-1190
- Cite (ACL):
- John Wieting and Kevin Gimpel. 2017. Revisiting Recurrent Networks for Paraphrastic Sentence Embeddings. In Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 2078–2088, Vancouver, Canada. Association for Computational Linguistics.
- Cite (Informal):
- Revisiting Recurrent Networks for Paraphrastic Sentence Embeddings (Wieting & Gimpel, ACL 2017)
- PDF:
- https://preview.aclanthology.org/nschneid-patch-1/P17-1190.pdf