CORE: A Conceptual Reasoning Layer for Large Language Models

TL;DR

CORE introduces a concept-first interaction layer for large language models that enhances multi-turn stability by maintaining a persistent semantic state, reducing prompt size and improving reasoning consistency.

Contribution

It proposes a novel, model-agnostic layer that separates reasoning from language generation using a persistent local concept, without altering model weights.

Findings

42% reduction in prompt tokens in prototype

Improves multi-turn stability and reasoning consistency

Separates conceptual reasoning from language generation

Abstract

Large language models handle single-turn generation well, but multi-turn interactions still require the model to reconstruct user intent and task state from an expanding token history because internal representations do not persist across turns. This token-first paradigm leads to drift, inconsistent reasoning modes, and growing prompts as conversations deepen. We propose CORE, a concept-first interaction layer that improves multi-turn stability without modifying model weights. CORE combines a small library of universal cognitive operators with a persistent Local Concept - a compact semantic state capturing the task, constraints, preferences, and intermediate results. Each model call receives only this concept state, the user's latest instruction, and the selected operator, eliminating the need to replay full history. A preliminary prototype simulating CORE's behavior shows about 42% reduction in cumulative prompt tokens, though this number reflects prototype conditions and should not be interpreted as a real-world performance estimate. CORE offers a model-agnostic mechanism that separates conceptual reasoning from language generation, suggesting a scalable direction for more stable multi-turn systems.