T-NER: An All-Round Python Library for Transformer-based Named Entity Recognition

TL;DR

T-NER is a comprehensive Python library designed for transformer-based NER that supports model fine-tuning, cross-domain and cross-lingual evaluation, and includes an interactive web app for qualitative analysis.

Contribution

The paper introduces T-NER, a versatile library for NER that enables detailed study of LM generalization and provides tools for practical and research-oriented evaluation.

Findings

In-domain NER performance is competitive across datasets.

Cross-domain generalization remains challenging.

Fine-tuning on combined datasets improves domain-specific learning.

Abstract

Language model (LM) pretraining has led to consistent improvements in many NLP downstream tasks, including named entity recognition (NER). In this paper, we present T-NER (Transformer-based Named Entity Recognition), a Python library for NER LM finetuning. In addition to its practical utility, T-NER facilitates the study and investigation of the cross-domain and cross-lingual generalization ability of LMs finetuned on NER. Our library also provides a web app where users can get model predictions interactively for arbitrary text, which facilitates qualitative model evaluation for non-expert programmers. We show the potential of the library by compiling nine public NER datasets into a unified format and evaluating the cross-domain and cross-lingual performance across the datasets. The results from our initial experiments show that in-domain performance is generally competitive across datasets. However, cross-domain generalization is challenging even with a large pretrained LM, which has nevertheless capacity to learn domain-specific features if fine-tuned on a combined dataset. To facilitate future research, we also release all our LM checkpoints via the Hugging Face model hub.