Large Language Models (LLMs) have recently made significant strides in
complex reasoning tasks through the Chain-of-Thought technique. Despite this
progress, their reasoning is often constrained by their intrinsic
understanding, lacking external insights. To address this, we propose
Exchange-of-Thought (EoT), a novel framework that enables cross-model
communication during problem-solving. Drawing inspiration from network
topology, EoT integrates four unique communication paradigms: Memory, Report,
Relay, and Debate. This paper delves into the communication dynamics and volume
associated with each paradigm. To counterbalance the risks of incorrect
reasoning chains, we implement a robust confidence evaluation mechanism within
these communications. Our experiments across diverse complex reasoning tasks
demonstrate that EoT significantly surpasses established baselines,
underscoring the value of external insights in enhancing LLM performance.
Furthermore, we show that EoT achieves these superior results in a
cost-effective manner, marking a promising advancement for efficient and
collaborative AI problem-solving.

通过链式推理技术，大型语言模型（LLMs）在复杂推理任务中取得了显著进展。然而，它们的推理常常受到内在理解的局限，缺乏外部洞察力。为了解决这个问题，我们提出了一种名为 “交换思路（EoT）” 的新框架，它在问题解决过程中实现了跨模型通信。EoT 从网络拓扑中汲取灵感，整合了四种独特的通信范式：记忆、报告、中继和辩论。本文深入探讨了每种范式所涉及的通信动态和通信量。为了平衡不正确推理链的风险，我们在这些通信中实现了一个健壮的置信度评估机制。我们对各种复杂推理任务进行的实验表明，EoT 明显超过了已有的基线方法，凸显了外部洞察力在增强 LLM 性能方面的价值。此外，我们还展示了 EoT 以一种具有成本效益的方式实现这些卓越结果，标志着高效而协同的人工智能问题解决的有望发展。