Towards Trustworthy AutoGrading of Short, Multi-lingual, Multi-type Answers

Abstract

Autograding short textual answers has become much more feasible due to the rise of NLP and the increased availability of question-answer pairs brought about by a shift to online education. Autograding performance is still inferior to human grading. The statistical and black-box nature of state-of-the-art machine learning models makes them untrustworthy, raising ethical concerns and limiting their practical utility. Furthermore, the evaluation of autograding is typically confined to small, monolingual datasets for a specific question type. This study uses a large dataset consisting of about 10 million question-answer pairs from multiple languages covering diverse fields such as math and language, and strong variation in question and answer syntax. We demonstrate the effectiveness of finetuning transformer models for autograding for such complex datasets. Our best hyperparameter-tuned model yields an accuracy of about 86.5%, comparable to the state-of-the-art models that are less general and more tuned to a specific type of question, subject, and language. More importantly, we address trust and ethical concerns. By involving humans in the autograding process, we show how to improve the accuracy of automatically graded answers, achieving accuracy equivalent to that of teaching assistants. We also show how teachers can effectively control the type of errors made by the system and how they can validate efficiently that the autograder's performance on individual exams is close to the expected performance.

Similarity-Based Content Scoring - How to Make S-BERT Keep Up With BERT

“…Gomaa and Fahmy (2019) use pretrained skip-thought vectors and learn a logistic classifier over the component-wise product and absolute difference vectors. Schneider et al (2022) report promising results on a notpublicly-available dataset by learning embeddings for question-answer-pairs and utilize cosine similarity as distance metric.…”