Underreporting of errors in NLG output, and what to do about it

van Miltenburg, Emiel; Clinciu, Miruna-Adriana; Du�ek, Ond?ej; Gkatzia, Dimitra; Inglis, Stephanie; Lepp�nen, Leo; Mahamood, Saad; Manning, Emma; Schoch, Stephanie; Thomson, Craig; Wen, Luou

Abstract

We observe a severe under-reporting of the different kinds of errors that Natural Language Generation systems make. This is a problem, because mistakes are an important indicator of where systems should still be improved. If authors only report overall performance metrics, the research community is left in the dark about the specific weaknesses that are exhibited by `state-of-the-art' research. Next to quantifying the extent of error under-reporting, this position paper provides recommendations for error identification, analysis and reporting.

• Presentation Conference Type: Conference Paper (Published)
• Conference Name: 14th International Conference on Natural Language Generation
• Start Date: Sep 20, 2021
• End Date: Sep 24, 2021
• Acceptance Date: Jul 26, 2021
• Publication Date: 2021
• Deposit Date: Aug 10, 2021
• Publicly Available Date: Aug 10, 2021
• Pages: 140-153
• Book Title: Proceedings of the 14th International Conference on Natural Language Generation
• Public URL: http://researchrepository.napier.ac.uk/Output/2791216
• Publisher URL: https://aclanthology.org/2021.inlg-1.14/
• Related Public URLs: https://inlg2021.github.io/pages/calls.html