Boundary Smoothing for Named Entity Recognition
This addresses performance and calibration problems in NER for NLP practitioners, but it is incremental as it builds on existing span-based methods.
The paper tackles the over-confidence issue in neural named entity recognition (NER) models by proposing boundary smoothing as a regularization technique, achieving results better than or competitive with previous state-of-the-art systems on eight NER benchmarks.
Neural named entity recognition (NER) models may easily encounter the over-confidence issue, which degrades the performance and calibration. Inspired by label smoothing and driven by the ambiguity of boundary annotation in NER engineering, we propose boundary smoothing as a regularization technique for span-based neural NER models. It re-assigns entity probabilities from annotated spans to the surrounding ones. Built on a simple but strong baseline, our model achieves results better than or competitive with previous state-of-the-art systems on eight well-known NER benchmarks. Further empirical analysis suggests that boundary smoothing effectively mitigates over-confidence, improves model calibration, and brings flatter neural minima and more smoothed loss landscapes.