Differentiable Prompt Makes Pre-trained Language Models Better Few-shot Learners

Large-scale pre-trained language models have contributed significantly to natural language processing by demonstrating remarkable abilities as few-shot learners. However, their effectiveness depends mainly on scaling the model parameters and prompt design, hindering their implementation in most real-world applications. This study proposes a novel pluggable, extensible, and efficient approach named DifferentiAble pRompT (DART), which can convert small language models into better few-shot learners without any prompt engineering. The main principle behind this approach involves reformulating potential natural language processing tasks into the task of a pre-trained language model and differentially optimizing the prompt template as well as the target label with backpropagation. Furthermore, the proposed approach can be: (i) Plugged to any pre-trained language models; (ii) Extended to widespread classification tasks. A comprehensive evaluation of standard NLP tasks demonstrates that the proposed approach achieves a better few-shot performance. Code is available in https://github.com/zjunlp/DART.

PDF Abstract ICLR 2022 PDF ICLR 2022 Abstract
Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Few-Shot Learning CR DART Acc 91.8(0.5) # 1
Few-Shot Learning GLUE QQP DART F1-score 67.8(3.2) # 1
Few-Shot Learning MR DART Acc 88.2(1.0) # 1
Few-Shot Learning MRPC DART F1-score 78.3(4.5) # 1
Few-Shot Learning SST-2 Binary classification DART Acc 93.5(0.5) # 1

Methods


No methods listed for this paper. Add relevant methods here