Making Pre-trained Language Models End-to-end Few-shot Learners with Contrastive Prompt Tuning

TL;DR

This paper introduces CP-Tuning, an end-to-end contrastive prompt tuning method that enables pre-trained language models to perform few-shot learning without manual prompt engineering, improving task-invariance and class distinction.

Contribution

It presents the first fully trainable, verbalizer-free contrastive prompt tuning framework with task-invariant continuous prompts and cost-sensitive contrastive learning.

Findings

CP-Tuning outperforms state-of-the-art methods on various language understanding tasks.

It effectively eliminates the need for manual prompt engineering.

The approach enhances class distinction and decision boundary smoothness.

Abstract

Pre-trained Language Models (PLMs) have achieved remarkable performance for various language understanding tasks in IR systems, which require the fine-tuning process based on labeled training data. For low-resource scenarios, prompt-based learning for PLMs exploits prompts as task guidance and turns downstream tasks into masked language problems for effective few-shot fine-tuning. In most existing approaches, the high performance of prompt-based learning heavily relies on handcrafted prompts and verbalizers, which may limit the application of such approaches in real-world scenarios. To solve this issue, we present CP-Tuning, the first end-to-end Contrastive Prompt Tuning framework for fine-tuning PLMs without any manual engineering of task-specific prompts and verbalizers. It is integrated with the task-invariant continuous prompt encoding technique with fully trainable prompt parameters. We further propose the pair-wise cost-sensitive contrastive learning procedure to optimize the model in order to achieve verbalizer-free class mapping and enhance the task-invariance of prompts. It explicitly learns to distinguish different classes and makes the decision boundary smoother by assigning different costs to easy and hard cases. Experiments over a variety of language understanding tasks used in IR systems and different PLMs show that CP-Tuning outperforms state-of-the-art methods.