We cast nested named entity recognition (NNER) as a sequence labeling task by leveraging prior work that linearizes constituency structures, effectively reducing the complexity of this structured prediction problem to straightforward token classification. By combining these constituency linearizations with pretrained encoders, our method captures nested entities while performing exactly n tagging actions. Our approach achieves competitive performance compared to less efficient systems, and it can be trained using any off-the-shelf sequence labeling library.
@article{arxiv.2505.16855,
title = {Nested Named Entity Recognition as Single-Pass Sequence Labeling},
author = {Alberto Muñoz-Ortiz and David Vilares and Caio Corro and Carlos Gómez-Rodríguez},
journal= {arXiv preprint arXiv:2505.16855},
year = {2025}
}