Entailment as Few-Shot Learner

• DOI: 10.48550/arxiv.2104.14690
• Publication Date: 2021-01-01
• Authors: Wang, Sinong; Fang, Han; Khabsa, Madian; Mao, Hanzi; Ma, Hao
• Keywords: FOS: Computer and information sciences; Computer Science - Computation and Language; Artificial Intelligence (cs.AI); Computer Science - Artificial Intelligence; 01 natural sciences; Computation and Language (cs.CL); 0105 earth and related environmental sciences

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 as 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: (i) naturally combined with an unsupervised contrastive learning-based data augmentation method; (ii) 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 times larger models, such as GPT-3.