Simple and Effective Paraphrastic Similarity from Parallel Translations
@inproceedings{Wieting2019SimpleAE, title={Simple and Effective Paraphrastic Similarity from Parallel Translations}, author={J. Wieting and Kevin Gimpel and Graham Neubig and Taylor Berg-Kirkpatrick}, booktitle={ACL}, year={2019} }
We present a model and methodology for learning paraphrastic sentence embeddings directly from bitext, removing the time-consuming intermediate step of creating para-phrase corpora. Further, we show that the resulting model can be applied to cross lingual tasks where it both outperforms and is orders of magnitude faster than more complex state-of-the-art baselines.
Supplemental Code
Github Repo
Via Papers with Code
Python code for training models in the ACL paper, "Simple and Effective Paraphrastic Similarity from Parallel Translations".
Figures, Tables, and Topics from this paper
20 Citations
Simulated Multiple Reference Training Improves Low-Resource Machine Translation
- Computer Science
- EMNLP
- 2020
- 4
- PDF
Paraphrase Generation as Zero-Shot Multilingual Translation: Disentangling Semantic Similarity from Lexical and Syntactic Diversity
- Computer Science
- WMT
- 2020
- 4
- PDF
A Deep Reinforced Model for Zero-Shot Cross-Lingual Summarization with Bilingual Semantic Similarity Rewards
- Computer Science
- NGT@ACL
- 2020
- Highly Influenced
- PDF
Automatic Machine Translation Evaluation in Many Languages via Zero-Shot Paraphrasing
- Computer Science
- EMNLP
- 2020
- 9
- PDF
Experiments on Paraphrase Identification Using Quora Question Pairs Dataset
- Computer Science
- ArXiv
- 2020
- 2
- PDF
References
SHOWING 1-10 OF 42 REFERENCES
ParaBank: Monolingual Bitext Generation and Sentential Paraphrasing via Lexically-constrained Neural Machine Translation
- Computer Science
- AAAI
- 2019
- 22
- PDF
Learning Joint Multilingual Sentence Representations with Neural Machine Translation
- Computer Science
- Rep4NLP@ACL
- 2017
- 96
- PDF
Pushing the Limits of Paraphrastic Sentence Embeddings with Millions of Machine Translations
- Computer Science, Mathematics
- ACL
- 2018
- 132
- PDF
An Empirical Analysis of NMT-Derived Interlingual Embeddings and Their Use in Parallel Sentence Identification
- Computer Science
- IEEE Journal of Selected Topics in Signal Processing
- 2017
- 42
- PDF
Jointly optimizing word representations for lexical and sentential tasks with the C-PHRASE model
- Computer Science
- ACL
- 2015
- 46
- PDF
Extracting Parallel Sentences with Bidirectional Recurrent Neural Networks to Improve Machine Translation
- Computer Science, Mathematics
- COLING
- 2018
- 22
- PDF