Interpreting Latent Student Knowledge Representations in Programming Assignments

arXiv:2405.08213v1 Announce Type: cross
Abstract: Recent advances in artificial intelligence for education leverage generative large language models, including using them to predict open-ended student responses rather than their correctness only. However, the black-box nature of these models limits the interpretability of the learned student knowledge representations. In this paper, we conduct a first exploration into interpreting latent student knowledge representations by presenting InfoOIRT, an Information regularized Open-ended Item Response Theory model, which encourages the latent student knowledge states to be …

abstract advances artificial artificial intelligence arxiv box cs.cl cs.cy cs.lg education generative however intelligence interpretability knowledge language language models large language large language models nature paper programming responses them type