Named Entity Recognition with stack residual LSTM and trainable bias decoding
This work improves NER performance for NLP applications, but it is incremental as it builds on existing RNN models.
The paper tackled Named Entity Recognition by introducing residual connections in stacked RNNs to address degradation and a bias decoding mechanism for optimizing non-differentiable objectives like F-measure, achieving state-of-the-art results on Spanish and English CoNLL 2003 datasets.
Recurrent Neural Network models are the state-of-the-art for Named Entity Recognition (NER). We present two innovations to improve the performance of these models. The first innovation is the introduction of residual connections between the Stacked Recurrent Neural Network model to address the degradation problem of deep neural networks. The second innovation is a bias decoding mechanism that allows the trained system to adapt to non-differentiable and externally computed objectives, such as the entity-based F-measure. Our work improves the state-of-the-art results for both Spanish and English languages on the standard train/development/test split of the CoNLL 2003 Shared Task NER dataset.