# Return of the RNN: Residual Recurrent Networks for Invertible Sentence Embeddings

arXiv: https://arxiv.org/abs/2303.13570 · HTML: https://mlsanity.com/paper/2303.13570/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 10 of its 18 references that are in the map.
- First posted on arXiv 23 Mar 2023.

## Builds on (most cited first, 10 of 10)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [Adam: A Method for Stochastic Optimization](https://mlsanity.com/paper/1412.6980/) (2014)
- [Neural Machine Translation by Jointly Learning to Align and Translate](https://mlsanity.com/paper/1409.0473/) (2014)
- [Efficient Estimation of Word Representations in Vector Space](https://mlsanity.com/paper/1301.3781/) (2013)
- [Learning Phrase Representations using RNN Encoder-Decoder for Statistical Machine Translation](https://mlsanity.com/paper/1406.1078/) (2014)
- [Distributed Representations of Words and Phrases and their Compositionality](https://mlsanity.com/paper/1310.4546/) (2013)
- [Sequence to Sequence Learning with Neural Networks](https://mlsanity.com/paper/1409.3215/) (2014)
- [Identity Mappings in Deep Residual Networks](https://mlsanity.com/paper/1603.05027/) (2016)
- [Skip-Thought Vectors](https://mlsanity.com/paper/1506.06726/) (2015)
- [An efficient framework for learning sentence representations](https://mlsanity.com/paper/1803.02893/) (2018)

## Abstract (arXiv)

This study presents a novel model for invertible sentence embeddings using a residual recurrent network trained on an unsupervised encoding task. Rather than the probabilistic outputs common to neural machine translation models, our approach employs a regression-based output layer to reconstruct the input sequence's word vectors. The model achieves high accuracy and fast training with the ADAM optimizer, a significant finding given that RNNs typically require memory units, such as LSTMs, or second-order optimization methods. We incorporate residual connections and introduce a "match drop" technique, where gradients are calculated only for incorrect words. Our approach demonstrates potential for various natural language processing applications, particularly in neural network-based systems that require high-quality sentence embeddings.