SMART: Robust and Efficient Fine-Tuning for Pre-trained Natural Language Models through Principled Regularized Optimization

By Haoming Jiang, Pengcheng He, Weizhu Chen, Xiaodong Liu, Jianfeng Gao, Tuo Zhao. 2019

613 citations recorded by the source corpus.

Transfer learning has fundamentally changed the landscape of natural language processing (NLP). Many state-of-the-art models are first pre-trained on a large text corpus and then fine-tuned on downstream tasks. However, due to limited data resources from downstream tasks and the extremely high complexity of pre-trained models, aggressive fine-tuning often causes the fine-tuned model to overfit the training data of downstream tasks and fail to generalize to unseen data. To address such an issue in a principled manner, we propose a new learning framework for robust and efficient fine-tuning for pre-trained models to attain better generalization performance. The proposed framework contains two important ingredients: 1. Smoothness-inducing regularization, which effectively manages the complexity of the model; 2. Bregman proximal point optimization, which is an instance of trust-region methods and can prevent aggressive updating. Our experiments show that the proposed framework achieves new state-of-the-art performance on a number of NLP tasks including GLUE, SNLI, SciTail and ANLI. Moreover, it also outperforms the state-of-the-art T5 model, which is the largest pre-trained model containing 11 billion parameters, on GLUE.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition, Alignment & RLHF, Optimization & Training

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

  1. Mohit Bansal
  2. Jianfeng Gao
  3. Luke S. Zettlemoyer
  4. Percy Liang
  5. Yejin Choi
  6. Hannaneh Hajishirzi
  7. Caiming Xiong
  8. Graham Neubig

In Computer Vision and Pattern Recognition

  1. Jianfeng Gao
  2. Luc Van Gool
  3. Dacheng Tao
  4. Mohit Bansal
  5. Trevor Darrell
  6. Xiaogang Wang
  7. Cordelia Schmid
  8. Ming-Hsuan Yang

In Alignment & RLHF

  1. Xiaodong Liu
  2. Jianfeng Gao
  3. Mari Ostendorf
  4. Noah A. Smith
  5. Luke S. Zettlemoyer
  6. Diyi Yang
  7. Hannaneh Hajishirzi
  8. Weijia Shi

In Optimization & Training

  1. Xiaodong Liu
  2. Jianfeng Gao
  3. Weizhu Chen
  4. Tian Li
  5. Virginia Smith
  6. Luke S. Zettlemoyer
  7. Mohit Bansal
  8. Pengcheng He

Citations in the corpus