Layered Chain-of-Thought Prompting for Multi-Agent LLM Systems: A Comprehensive Approach to Explainable Large Language Models

TL;DR

This paper introduces Layered-CoT prompting, a multi-layered reasoning framework for large language models that enhances transparency, correctness, and user engagement in complex, high-stakes tasks.

Contribution

It proposes a novel layered reasoning approach with external verification and user feedback, improving explainability and reliability over traditional chain-of-thought prompting.

Findings

Layered-CoT improves transparency and correctness.

It enhances user engagement in reasoning processes.

Demonstrated effectiveness in medical, financial, and engineering scenarios.

Abstract

Large Language Models (LLMs) leverage chain-of-thought (CoT) prompting to provide step-by-step rationales, improving performance on complex tasks. Despite its benefits, vanilla CoT often fails to fully verify intermediate inferences and can produce misleading explanations. In this work, we propose Layered Chain-of-Thought (Layered-CoT) Prompting, a novel framework that systematically segments the reasoning process into multiple layers, each subjected to external checks and optional user feedback. We expand on the key concepts, present three scenarios -- medical triage, financial risk assessment, and agile engineering -- and demonstrate how Layered-CoT surpasses vanilla CoT in terms of transparency, correctness, and user engagement. By integrating references from recent arXiv papers on interactive explainability, multi-agent frameworks, and agent-based collaboration, we illustrate how Layered-CoT paves the way for more reliable and grounded explanations in high-stakes domains.