papers_we_read

Siamese Recurrent Architectures for Learning Sentence Similarity

Jonas Mueller, Aditya Thyagarajan, AAAI-2016

Summary

This paper proposes a siamese adaptation of the Long Short-Term Memory [LSTM] network for labeled data comprised of pairs of variable-length sequences. The proposed model is applied to assess semantic similarity between sentences, improving over the state of the art, outperforming carefully handcrafted features and recently proposed neural network systems of greater complexity.

Main contributions:

Strengths

Weaknesses / Notes

Implementation

https://github.com/dhwajraj/deep-siamese-text-similarity/
https://github.com/aditya1503/Siamese-LSTM