Evaluating the Robustness of Neural Language Models to Input Perturbations

by   Milad Moradi, et al.

High-performance neural language models have obtained state-of-the-art results on a wide range of Natural Language Processing (NLP) tasks. However, results for common benchmark datasets often do not reflect model reliability and robustness when applied to noisy, real-world data. In this study, we design and implement various types of character-level and word-level perturbation methods to simulate realistic scenarios in which input texts may be slightly noisy or different from the data distribution on which NLP systems were trained. Conducting comprehensive experiments on different NLP tasks, we investigate the ability of high-performance language models such as BERT, XLNet, RoBERTa, and ELMo in handling different types of input perturbations. The results suggest that language models are sensitive to input perturbations and their performance can decrease even when small changes are introduced. We highlight that models need to be further improved and that current benchmarks are not reflecting model robustness well. We argue that evaluations on perturbed inputs should routinely complement widely-used benchmarks in order to yield a more realistic understanding of NLP systems robustness.


page 1

page 2

page 3

page 4


Deep learning models are not robust against noise in clinical text

Artificial Intelligence (AI) systems are attracting increasing interest ...

Improving the robustness and accuracy of biomedical language models through adversarial training

Deep transformer neural network models have improved the predictive accu...

Improving Robustness and Generality of NLP Models Using Disentangled Representations

Supervised neural networks, which first map an input x to a single repre...

How Robust Are Character-Based Word Embeddings in Tagging and MT Against Wrod Scramlbing or Randdm Nouse?

This paper investigates the robustness of NLP against perturbed word for...

Is the Computation of Abstract Sameness Relations Human-Like in Neural Language Models?

In recent years, deep neural language models have made strong progress i...

Reliability Testing for Natural Language Processing Systems

Questions of fairness, robustness, and transparency are paramount to add...

Evaluating NLP Systems On a Novel Cloze Task: Judging the Plausibility of Possible Fillers in Instructional Texts

Cloze task is a widely used task to evaluate an NLP system's language un...