KARL: Knowledge-Aware Retrieval and Representations aid Retention and Learning in Students

This paper presents KAR3L, a content-aware flashcard scheduling system that uses deep knowledge tracing, BERT retrieval, and semantic understanding of flashcard content to predict student recall and optimize learning sequences. The system was evaluated with 27 users studying trivia questions, showing improved testing throughput over existing methods.

Flashcard schedulers rely on 1) *student models* to predict the flashcards a student knows; and 2) *teaching policies* to pick which cards to show next via these predictions.Prior student models, however, just use study data like the student’s past responses, ignoring the text on cards. We propose **content-aware scheduling**, the first schedulers exploiting flashcard content.To give the first evidence that such schedulers enhance student learning, we build KARL, a simple but effective content-a

Source

Framework Categories

Tool Types

Tags