Seq2Biseq: Bidirectional Output-wise Recurrent Neural Networks for Sequence Modelling

TL;DR

Seq2Biseq introduces a bidirectional RNN architecture with gated layers and dual decoders for improved sequence labeling, demonstrating competitive or superior performance and scalability to large datasets.

Contribution

The paper presents a novel bidirectional RNN architecture with two decoders for sequence labeling, enhancing context capture and performance.

Findings

Achieves results better than or close to state-of-the-art.

Scales effectively to larger corpora.

Utilizes recent technologies for improved performance.

Abstract

During the last couple of years, Recurrent Neural Networks (RNN) have reached state-of-the-art performances on most of the sequence modelling problems. In particular, the "sequence to sequence" model and the neural CRF have proved to be very effective in this domain. In this article, we propose a new RNN architecture for sequence labelling, leveraging gated recurrent layers to take arbitrarily long contexts into account, and using two decoders operating forward and backward. We compare several variants of the proposed solution and their performances to the state-of-the-art. Most of our results are better than the state-of-the-art or very close to it and thanks to the use of recent technologies, our architecture can scale on corpora larger than those used in this work.