Context-aware Adversarial Attack on Named Entity Recognition

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Chen, Shuguang, Neves, Leonardo, Solorio, Thamar
Natura: Preprint
Pubblicazione: 2023
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866914664500166656
author Chen, Shuguang
Neves, Leonardo
Solorio, Thamar
author_facet Chen, Shuguang
Neves, Leonardo
Solorio, Thamar
contents In recent years, large pre-trained language models (PLMs) have achieved remarkable performance on many natural language processing benchmarks. Despite their success, prior studies have shown that PLMs are vulnerable to attacks from adversarial examples. In this work, we focus on the named entity recognition task and study context-aware adversarial attack methods to examine the model's robustness. Specifically, we propose perturbing the most informative words for recognizing entities to create adversarial examples and investigate different candidate replacement methods to generate natural and plausible adversarial examples. Experiments and analyses show that our methods are more effective in deceiving the model into making wrong predictions than strong baselines.
format Preprint
id arxiv_https___arxiv_org_abs_2309_08999
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Context-aware Adversarial Attack on Named Entity Recognition
Chen, Shuguang
Neves, Leonardo
Solorio, Thamar
Computation and Language
In recent years, large pre-trained language models (PLMs) have achieved remarkable performance on many natural language processing benchmarks. Despite their success, prior studies have shown that PLMs are vulnerable to attacks from adversarial examples. In this work, we focus on the named entity recognition task and study context-aware adversarial attack methods to examine the model's robustness. Specifically, we propose perturbing the most informative words for recognizing entities to create adversarial examples and investigate different candidate replacement methods to generate natural and plausible adversarial examples. Experiments and analyses show that our methods are more effective in deceiving the model into making wrong predictions than strong baselines.
title Context-aware Adversarial Attack on Named Entity Recognition
topic Computation and Language
url https://arxiv.org/abs/2309.08999