Investigating Transformers for Automatic Short Answer Grading

Recent advancements in the field of deep learning for natural language processing made it possible to use novel deep learning architectures, such as the Transformer, for increasingly complex natural language processing tasks. Combined with novel unsupervised pre-training tasks such as masked language modeling, sentence ordering or next sentence prediction, those natural language processing models became even more accurate. In this work, we experiment with fine-tuning different pretrained Transformer based architectures. We train the newest and most powerful, according to the glue benchmark, transformers on the SemEval2013 dataset. We also explore the impact of transfer learning a model fine-tuned on the MNLI dataset to the SemEval-2013 dataset on generalization and performance. We report up to 13% absolute improvement in macro-average-F1 over state-of-the-art results. We show that models trained with knowledge distillation are feasible for use in short answer grading. Furthermore, we compare multilingual models on a machinetranslated version of the SemEval-2013 dataset.