paper-with-me

Papers

Evaluating the Robustness of Neural Language Models to Input Perturbations

2021-08-27 · EMNLP 2021 11 · Milad Moradi, Matthias Samwald

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.

📄 PDF Abstract BibTeX arXiv:2108.12237

Code (1)

mmoradi-iut/nlp-perturbation 공식 구현

Methods 이 논문이 사용한 방법론

Attention 설명 없음
Linear Layer A Linear Layer is a projection $\mathbf{XW + b}$.
Sigmoid Activation 설명 없음
Tanh Activation 설명 없음
LSTM An LSTM is a type of recurrent neural network that addresses the vanishing gradient problem in vanilla…
BiLSTM A Bidirectional LSTM, or biLSTM, is a sequence processing model that consists of two LSTMs: one taking the input in a forward direction, and the other in a backwards…
Dropout Dropout is a regularization technique for neural networks that drops a unit (along with connections) at training time with a specified probability $p$ (a common value is…
Adam 설명 없음

Similar Papers 제목 키워드 기반

Evaluating Robustness to Input Perturbations for Neural Machine Translation

2020-05-01 · ACL 2020 6 · Xing Niu, Prashant Mathur, Georgiana Dinu, Yaser Al-Onaizan

Neural Machine Translation (NMT) models are sensitive to small perturbations in the input. Robustness to such perturbations is typically measured using translation quality metrics such as BLEU on the noisy input. This pa…

Machine TranslationNMTTranslation

Evaluating Concurrent Robustness of Language Models Across Diverse Challenge Sets

2023-11-15 · Vatsal Gupta, Pranshu Pandya, Tushar Kataria, Vivek Gupta 외

Language models, characterized by their black-box nature, often hallucinate and display sensitivity to input perturbations, causing concerns about trust. To enhance trust, it is imperative to gain a comprehensive underst…

Revisiting DeepFool: generalization and improvement

2023-03-22 · Alireza Abdollahpoorrostam, Mahed Abroshan, Seyed-Mohsen Moosavi-Dezfooli

Deep neural networks have been known to be vulnerable to adversarial examples, which are inputs that are modified slightly to fool the network into making incorrect predictions. This has led to a significant amount of re…

Adversarial AttackAdversarial RobustnessComputational Efficiency

Interpretability Illusions with Sparse Autoencoders: Evaluating Robustness of Concept Representations

2025-05-21 · Aaron J. Li, Suraj Srinivas, Usha Bhalla, Himabindu Lakkaraju

Sparse autoencoders (SAEs) are commonly used to interpret the internal activations of large language models (LLMs) by mapping them to human-interpretable concept representations. While existing evaluations of SAEs focus …

Disentanglement

Stochasticity in Tokenisation Improves Robustness

2026-04-17 · Sophie Steger, Rui Li, Sofiane Ennadir, Anya Sims 외 arxiv

The widespread adoption of large language models (LLMs) has increased concerns about their robustness. Vulnerabilities in perturbations of tokenisation of the input indicate that models trained with a deterministic canon…