Improving and Simplifying Pattern Exploiting Training

TL;DR

This paper introduces ADAPET, an improved pattern exploiting training method that enhances few-shot learning performance of pre-trained language models without requiring task-specific unlabeled data.

Contribution

ADAPET modifies PET's objective to provide denser supervision, enabling better few-shot learning performance without unlabeled data.

Findings

ADAPET outperforms PET on SuperGLUE benchmarks.

ADAPET achieves strong results without task-specific unlabeled data.

The method simplifies pattern-based training for language models.

Abstract

Recently, pre-trained language models (LMs) have achieved strong performance when fine-tuned on difficult benchmarks like SuperGLUE. However, performance can suffer when there are very few labeled examples available for fine-tuning. Pattern Exploiting Training (PET) is a recent approach that leverages patterns for few-shot learning. However, PET uses task-specific unlabeled data. In this paper, we focus on few-shot learning without any unlabeled data and introduce ADAPET, which modifies PET's objective to provide denser supervision during fine-tuning. As a result, ADAPET outperforms PET on SuperGLUE without any task-specific unlabeled data. Our code can be found at https://github.com/rrmenon10/ADAPET.