IM-BERT: Enhancing Robustness of BERT through the Implicit Euler Method

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Kim, Mihyeon, Park, Juhyoung, Kim, Youngbin
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915282793005056
author Kim, Mihyeon
Park, Juhyoung
Kim, Youngbin
author_facet Kim, Mihyeon
Park, Juhyoung
Kim, Youngbin
contents Pre-trained Language Models (PLMs) have achieved remarkable performance on diverse NLP tasks through pre-training and fine-tuning. However, fine-tuning the model with a large number of parameters on limited downstream datasets often leads to vulnerability to adversarial attacks, causing overfitting of the model on standard datasets. To address these issues, we propose IM-BERT from the perspective of a dynamic system by conceptualizing a layer of BERT as a solution of Ordinary Differential Equations (ODEs). Under the situation of initial value perturbation, we analyze the numerical stability of two main numerical ODE solvers: the explicit and implicit Euler approaches. Based on these analyses, we introduce a numerically robust IM-connection incorporating BERT's layers. This strategy enhances the robustness of PLMs against adversarial attacks, even in low-resource scenarios, without introducing additional parameters or adversarial training strategies. Experimental results on the adversarial GLUE (AdvGLUE) dataset validate the robustness of IM-BERT under various conditions. Compared to the original BERT, IM-BERT exhibits a performance improvement of approximately 8.3\%p on the AdvGLUE dataset. Furthermore, in low-resource scenarios, IM-BERT outperforms BERT by achieving 5.9\%p higher accuracy.
format Preprint
id arxiv_https___arxiv_org_abs_2505_06889
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle IM-BERT: Enhancing Robustness of BERT through the Implicit Euler Method
Kim, Mihyeon
Park, Juhyoung
Kim, Youngbin
Computation and Language
Artificial Intelligence
Pre-trained Language Models (PLMs) have achieved remarkable performance on diverse NLP tasks through pre-training and fine-tuning. However, fine-tuning the model with a large number of parameters on limited downstream datasets often leads to vulnerability to adversarial attacks, causing overfitting of the model on standard datasets. To address these issues, we propose IM-BERT from the perspective of a dynamic system by conceptualizing a layer of BERT as a solution of Ordinary Differential Equations (ODEs). Under the situation of initial value perturbation, we analyze the numerical stability of two main numerical ODE solvers: the explicit and implicit Euler approaches. Based on these analyses, we introduce a numerically robust IM-connection incorporating BERT's layers. This strategy enhances the robustness of PLMs against adversarial attacks, even in low-resource scenarios, without introducing additional parameters or adversarial training strategies. Experimental results on the adversarial GLUE (AdvGLUE) dataset validate the robustness of IM-BERT under various conditions. Compared to the original BERT, IM-BERT exhibits a performance improvement of approximately 8.3\%p on the AdvGLUE dataset. Furthermore, in low-resource scenarios, IM-BERT outperforms BERT by achieving 5.9\%p higher accuracy.
title IM-BERT: Enhancing Robustness of BERT through the Implicit Euler Method
topic Computation and Language
Artificial Intelligence
url https://arxiv.org/abs/2505.06889