Cognitive Agent Compilation for Explicit Problem Solver Modeling

TL;DR

This paper introduces Cognitive Agent Compilation (CAC), a framework that transforms large language models into explicit, inspectable problem-solving agents suitable for educational use, emphasizing transparency and editability.

Contribution

It proposes a novel method to compile LLM knowledge into explicit, controllable agents, addressing transparency and editability in educational AI systems.

Findings

CAC surfaces key design trade-offs between explicit control and scalability.

Proof of concept with Small Language Models demonstrates feasibility.

Framework enhances inspectability and editability of problem-solving knowledge.

Abstract

Large language models (LLMs) are widely used for tutoring, feedback generation, and content creation, but their broad pretraining makes them hard to constrain and poor substitutes for controllable learners. Educational systems often require inspectable and editable knowledge states: educators want to know what a system assumes the learner knows, and learners benefit when the system can justify actions in terms of explicit skills, misconceptions, and strategies. Inspired by cognitive architectures, we propose Cognitive Agent Compilation (CAC), a framework that uses a strong teacher LLM to compile problem-solving knowledge into an explicit target agent. CAC separates (i) knowledge representation, (ii) problem-solving policy, and (iii) verification and update rules, with the goal of making bounded problem solving more inspectable and editable in educational settings. We present an early proof of concept implemented with Small Language Models that surfaces key design trade-offs, particularly between explicit control and scalable generalization, and positions CAC as an initial step toward bounded-knowledge AI for educational applications.