Improving Pre-trained Language Model Fine-tuning with Noise Stability Regularization
This addresses the problem of poor generalizability in fine-tuning for NLP practitioners, offering an incremental improvement over existing regularization techniques.
The paper tackles overfitting in fine-tuning pre-trained language models by proposing Layerwise Noise Stability Regularization (LNSR), which injects noise into hidden representations and improves performance on text classification and question answering tasks, achieving state-of-the-art results over methods like L2-SP and SMART.
The advent of large-scale pre-trained language models has contributed greatly to the recent progress in natural language processing. Many state-of-the-art language models are first trained on a large text corpus and then fine-tuned on downstream tasks. Despite its recent success and wide adoption, fine-tuning a pre-trained language model often suffers from overfitting, which leads to poor generalizability due to the extremely high complexity of the model and the limited training samples from downstream tasks. To address this problem, we propose a novel and effective fine-tuning framework, named Layerwise Noise Stability Regularization (LNSR). Specifically, we propose to inject the standard Gaussian noise or In-manifold noise and regularize hidden representations of the fine-tuned model. We first provide theoretical analyses to support the efficacy of our method. We then demonstrate the advantages of the proposed method over other state-of-the-art algorithms including L2-SP, Mixout and SMART. While these previous works only verify the effectiveness of their methods on relatively simple text classification tasks, we also verify the effectiveness of our method on question answering tasks, where the target problem is much more difficult and more training examples are available. Furthermore, extensive experimental results indicate that the proposed algorithm can not only enhance the in-domain performance of the language models but also improve the domain generalization performance on out-of-domain data.