Towards Adaptive Pedagogical Policies: A Hybrid Reinforcement Learning and Large Language Model Framework for Intelligent Tutoring Systems under Indonesia's Curriculum

This vision paper proposes a hybrid architecture integrating Q-learning as a pedagogical decision engine and LLMs as a communication engine, orchestrated within an Adaptive Learning System (ALS) layer. The framework aims to support Indonesia’s Merdeka Curriculum by enabling adaptive pedagogical policies that are both strategic and conversational. Contributions include a principled integration of RL and LLM, localization to Indonesian secondary education, and a four-phase research roadmap. The paper avoids implementation details, focusing instead on conceptual clarity, curricular relevance, and dual-axis evaluation.