Explainable Automatic Grading with Neural Additive Models

arXiv:2405.00489v1 Announce Type: new
Abstract: The use of automatic short answer grading (ASAG) models may help alleviate the time burden of grading while encouraging educators to frequently incorporate open-ended items in their curriculum. However, current state-of-the-art ASAG models are large neural networks (NN) often described as "black box", providing no explanation for which characteristics of an input are important for the produced output. This inexplicable nature can be frustrating to teachers and students when trying to interpret, or learn from …

abstract art arxiv black box box cs.cl cs.lg current curriculum however networks neural networks stat.ap state type while