Distinguish Before Answer: Generating Contrastive Explanation as Knowledge for Commonsense Question Answering
This addresses the challenge of enhancing commonsense reasoning in AI systems, though it is incremental as it builds on existing knowledge-enhanced methods.
The paper tackles the problem of limited knowledge relevance and distinguishment in commonsense question answering by proposing CPACE, a model that generates contrastive explanations from symbolic knowledge, achieving new state-of-the-art results with 89.8% accuracy on CSQA (0.9% higher than human performance) and improvements of 4.2% and 3.5% on QASC and OBQA.
Existing knowledge-enhanced methods have achieved remarkable results in certain QA tasks via obtaining diverse knowledge from different knowledge bases. However, limited by the properties of retrieved knowledge, they still have trouble benefiting from both the knowledge relevance and distinguishment simultaneously. To address the challenge, we propose CPACE, a Concept-centric Prompt-bAsed Contrastive Explanation Generation model, which aims to convert obtained symbolic knowledge into a contrastive explanation for better distinguishing the differences among given candidates. Firstly, following previous works, we retrieve different types of symbolic knowledge with a concept-centric knowledge extraction module. After that, we generate corresponding contrastive explanations using acquired symbolic knowledge and explanation prompts as guidance for better modeling the knowledge distinguishment and interpretability. Finally, we regard the generated contrastive explanation as external knowledge for downstream task enhancement. We conduct a series of experiments on three widely-used question-answering datasets: CSQA, QASC, and OBQA. Experimental results demonstrate that with the help of generated contrastive explanation, our CPACE model achieves new SOTA on CSQA (89.8% on the testing set, 0.9% higher than human performance), and gains impressive improvement on QASC and OBQA (4.2% and 3.5%, respectively).