It's Common Sense, isn't it? Demystifying Human Evaluations in Commonsense-enhanced NLG systems

Mahamood, Saad; Clinciu, Miruna; Gkatzia, Dimitra

Abstract

Common sense is an integral part of human cognition which allows us to make sound decisions , communicate effectively with others and interpret situations and utterances. Endowing AI systems with commonsense knowledge capabilities will help us get closer to creating systems that exhibit human intelligence. Recent efforts in Natural Language Generation (NLG) have focused on incorporating com-monsense knowledge through large-scale pre-trained language models or by incorporating external knowledge bases. Such systems exhibit reasoning capabilities without common sense being explicitly encoded in the training set. These systems require careful evaluation, as they incorporate additional resources during training which adds additional sources of errors. Additionally, human evaluation of such systems can have significant variation, making it impossible to compare different systems and define baselines. This paper aims to de-mystify human evaluations of commonsense-enhanced NLG systems by proposing the Com-monsense Evaluation Card (CEC), a set of recommendations for evaluation reporting of commonsense-enhanced NLG systems, underpinned by an extensive analysis of human evaluations reported in the recent literature.

• Presentation Conference Type: Conference Paper (Published)
• Conference Name: Workshop on Human Evaluation of NLP Systems (HumEval at EACL 2021)
• Start Date: Apr 19, 2021
• End Date: Apr 19, 2021
• Acceptance Date: Mar 22, 2021
• Publication Date: 2021-04
• Deposit Date: Apr 9, 2021
• Publicly Available Date: Apr 9, 2021
• Book Title: Proceedings of the Workshop on Human Evaluation of NLP Systems (HumEval)
• Public URL: http://researchrepository.napier.ac.uk/Output/2760100
• Publisher URL: https://aclanthology.org/2021.humeval-1.1