Context-aware Adversarial Training for Name Regularity Bias in Named Entity Recognition

In this work, we examine the ability of NER models to use contextual information when predicting the type of an ambiguous entity. We introduce NRB, a new testbed carefully designed to diagnose Name Regularity Bias of NER models. Our results indicate that all state-of-the-art models we tested show su...

Full description

Saved in:
Bibliographic Details
Published in:arXiv.org 2021-07
Main Authors: Ghaddar, Abbas, Langlais, Philippe, Rashid, Ahmad, Rezagholizadeh, Mehdi
Format: Article
Language:English
Subjects:
Bias
Freezing
Regularity
Training
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:In this work, we examine the ability of NER models to use contextual information when predicting the type of an ambiguous entity. We introduce NRB, a new testbed carefully designed to diagnose Name Regularity Bias of NER models. Our results indicate that all state-of-the-art models we tested show such a bias; BERT fine-tuned models significantly outperforming feature-based (LSTM-CRF) ones on NRB, despite having comparable (sometimes lower) performance on standard benchmarks. To mitigate this bias, we propose a novel model-agnostic training method that adds learnable adversarial noise to some entity mentions, thus enforcing models to focus more strongly on the contextual signal, leading to significant gains on NRB. Combining it with two other training strategies, data augmentation and parameter freezing, leads to further gains.
ISSN:2331-8422
DOI:10.48550/arxiv.2107.11610