TSGP: Two-Stage Generative Prompting for Unsupervised Commonsense Question Answering

TL;DR

This paper introduces TSGP, a two-stage prompt-based framework that leverages pre-trained language models' implicit knowledge to improve unsupervised commonsense question answering across multiple tasks.

Contribution

The paper proposes a novel two-stage prompting method that enhances unsupervised commonsense reasoning by generating knowledge and answers without task-specific labeled data.

Findings

Significant improvement on CommonsenseQA, OpenBookQA, and SocialIQA.

Effective utilization of implicit knowledge in pre-trained language models.

Outperforms previous unsupervised approaches in commonsense QA.

Abstract

Unsupervised commonsense question answering requires mining effective commonsense knowledge without the rely on the labeled task data. Previous methods typically retrieved from traditional knowledge bases or used pre-trained language models (PrLMs) to generate fixed types of knowledge, which have poor generalization ability. In this paper, we aim to address the above limitation by leveraging the implicit knowledge stored in PrLMs and propose a two-stage prompt-based unsupervised commonsense question answering framework (TSGP). Specifically, we first use knowledge generation prompts to generate the knowledge required for questions with unlimited types and possible candidate answers independent of specified choices. Then, we further utilize answer generation prompts to generate possible candidate answers independent of specified choices. Experimental results and analysis on three different commonsense reasoning tasks, CommonsenseQA, OpenBookQA, and SocialIQA, demonstrate that TSGP significantly improves the reasoning ability of language models in unsupervised settings. Our code is available at: https://github.com/Yueqing-Sun/TSGP.