Adversarial Training for Large Neural Language Models

by   Xiaodong Liu, et al.

Generalization and robustness are both key desiderata for designing machine learning methods. Adversarial training can enhance robustness, but past work often finds it hurts generalization. In natural language processing (NLP), pre-training large neural language models such as BERT have demonstrated impressive gain in generalization for a variety of tasks, with further improvement from adversarial fine-tuning. However, these models are still vulnerable to adversarial attacks. In this paper, we show that adversarial pre-training can improve both generalization and robustness. We propose a general algorithm ALUM (Adversarial training for large neural LangUage Models), which regularizes the training objective by applying perturbations in the embedding space that maximizes the adversarial loss. We present the first comprehensive study of adversarial training in all stages, including pre-training from scratch, continual pre-training on a well-trained model, and task-specific fine-tuning. ALUM obtains substantial gains over BERT on a wide range of NLP tasks, in both regular and adversarial scenarios. Even for models that have been well trained on extremely large text corpora, such as RoBERTa, ALUM can still produce significant gains from continual pre-training, whereas conventional non-adversarial methods can not. ALUM can be further combined with task-specific fine-tuning to attain additional gains. The ALUM code and pre-trained models will be made publicly available at


page 1

page 2

page 3

page 4


Task-specific Objectives of Pre-trained Language Models for Dialogue Adaptation

Pre-trained Language Models (PrLMs) have been widely used as backbones i...

MAT: Mixed-Strategy Game of Adversarial Training in Fine-tuning

Fine-tuning large-scale pre-trained language models has been demonstrate...

Impact of Adversarial Training on Robustness and Generalizability of Language Models

Adversarial training is widely acknowledged as the most effective defens...

Adversarial Self-Attention for Language Understanding

An ultimate language system aims at the high generalization and robustne...

Using Pre-Training Can Improve Model Robustness and Uncertainty

Tuning a pre-trained network is commonly thought to improve data efficie...

Recyclable Tuning for Continual Pre-training

Continual pre-training is the paradigm where pre-trained language models...

Enhance the Visual Representation via Discrete Adversarial Training

Adversarial Training (AT), which is commonly accepted as one of the most...

Please sign up or login with your details

Forgot password? Click here to reset