Sparse-Checklist Prompting for Arabic Grammar Tutoring: Fast, Token-Efficient Feedback

Zayan Hasan

Keywords: Token-efficient prompting, Arabic grammar tutoring, Reasoning cost, LLMs

TL;DR: We show that constraining LLM tutoring feedback to a small checklist of hint tags cuts latency and token cost by ~50% while improving accuracy in Arabic grammar exercises.

Abstract: We explore token-efficient prompting for Arabic grammar tutoring, where time and cost-efficient approaches to feedback are important for Muslim community classes. Rather than producing free-form explanations, we restrict the model to providing a single pedagogical hint tag from a set of 5 possible tags, (Sparse-Checklist), and implement a simple router that sends clearly correct outputs down a short path. On 180 items with skill-labeled responses in the categories of agreement, pronoun clitics, prepositions and definiteness, Sparse-Checklist enhanced correctness over a Direct feedback baseline (81.1\% versus 76.1\%), reduced median latency (0.530s versus 0.807s) and half the completion tokens, which we consider a realization of reasoning cost (11.9 versus 22.7). A combined Router variant achieves 79.4\% accuracy, while achieving 18.2 completion tokens and 0.639s median latency. On incorrect responses, Sparse-Checklist and Router both select the appropriate skill tag 100\% of the time.

Submission Number: 86