“Entailment As Few-Shot Learner”, Sinong Wang, Han Fang, Madian Khabsa, Hanzi Mao, Hao Ma2021-04-29 (; similar)⁠:

Large pre-trained language models (LMs) have demonstrated remarkable ability as few-shot learners. However, their success hinges largely on scaling model parameters to a degree that makes it challenging to train and serve.

In this paper, we propose a new approach, named EFL, that can turn small LMs into better few-shot learners. The key idea of this approach is to reformulate potential NLP task into an entailment one, and then fine-tune the model with as little as 8 examples. We further demonstrate our proposed method can be: (1) naturally combined with an unsupervised contrastive learning-based data augmentation method; (2) easily extended to multilingual few-shot learning.

A systematic evaluation on 18 standard NLP tasks demonstrates that this approach improves the various existing SOTA few-shot learning methods by 12%, and yields competitive few-shot performance with 500× larger models, such as GPT-3.