Benchmarking Knowledge-Enhanced Commonsense Question Answering via Knowledge-to-Text Transformation

TL;DR

This paper benchmarks knowledge-enhanced commonsense question answering using a knowledge-to-text framework, showing it achieves state-of-the-art results and highlighting unexplored potential and promising future directions.

Contribution

It introduces a simple knowledge-to-text transformation framework that sets a new performance baseline and analyzes the potential and future directions for knowledge-enhanced CQA.

Findings

Our framework achieves state-of-the-art on CommonsenseQA.

Significant gap remains between current models and models with golden knowledge.

Future promising directions include context-sensitive knowledge selection and rich language models.

Abstract

A fundamental ability of humans is to utilize commonsense knowledge in language understanding and question answering. In recent years, many knowledge-enhanced Commonsense Question Answering (CQA) approaches have been proposed. However, it remains unclear: (1) How far can we get by exploiting external knowledge for CQA? (2) How much potential of knowledge has been exploited in current CQA models? (3) Which are the most promising directions for future CQA? To answer these questions, we benchmark knowledge-enhanced CQA by conducting extensive experiments on multiple standard CQA datasets using a simple and effective knowledge-to-text transformation framework. Experiments show that: (1) Our knowledge-to-text framework is effective and achieves state-of-the-art performance on CommonsenseQA dataset, providing a simple and strong knowledge-enhanced baseline for CQA; (2) The potential of knowledge is still far from being fully exploited in CQA -- there is a significant performance gap from current models to our models with golden knowledge; and (3) Context-sensitive knowledge selection, heterogeneous knowledge exploitation, and commonsense-rich language models are promising CQA directions.