ConsPrompt: Exploiting Contrastive Samples for Fewshot Prompt Learning

TL;DR

ConsPrompt introduces a contrastive learning approach that leverages contrastive samples to enhance prompt robustness and performance in few-shot learning scenarios with pre-trained language models.

Contribution

It proposes a novel multi-degree contrastive learning framework for prompt-based fine-tuning, improving robustness and achieving state-of-the-art results in few-shot tasks.

Findings

State-of-the-art performance in various few-shot settings

Effective utilization of multi-degree contrastive learning

Validated through ablation experiments

Abstract

The prompt has become an effective linguistic tool for utilizing pre-trained language models. However, in few-shot scenarios, subtle changes in the prompt design always make the result widely different, and the prompt learning methods also make it easy to overfit the limited samples. To alleviate this, we explore utilizing suitable contrastive samples and multi-degree contrastive learning methods to improve the robustness of the prompt representation. Therefore, the proposed Consprompt combined with the prompt encoding network, contrastive sampling modules, and contrastive scoring modules, is introduced to realize differential contrastive learning. Our results exhibit state-of-the-art performance in different few-shot settings, and the ablation experiments also certify the effectiveness of utilizing multi-degree contrastive learning in the prompt-based fine-tuning process.