A Span Selection Model for Semantic Role Labeling
This improves SRL accuracy for NLP applications, but it is incremental as it builds on existing span-based approaches.
The paper tackles semantic role labeling by proposing a span-based model that scores all possible argument spans and uses greedy selection, achieving state-of-the-art results of 87.4 F1 on CoNLL-2005 and 87.0 F1 on CoNLL-2012.
We present a simple and accurate span-based model for semantic role labeling (SRL). Our model directly takes into account all possible argument spans and scores them for each label. At decoding time, we greedily select higher scoring labeled spans. One advantage of our model is to allow us to design and use span-level features, that are difficult to use in token-based BIO tagging approaches. Experimental results demonstrate that our ensemble model achieves the state-of-the-art results, 87.4 F1 and 87.0 F1 on the CoNLL-2005 and 2012 datasets, respectively.