Synergy-of-Thoughts：在混合语言模型中引发高效推理
Synergy-of-Thoughts: Eliciting Efficient Reasoning in Hybrid Language Models
AI总结：
受双过程理论启发，本文提出SoT框架，利用小模型生成直觉思维并交叉评估，冲突时调用大模型反思纠正，在保证准确率与多样性的同时大幅降低API成本。
摘要：
大型语言模型（LLMs）在广泛任务中展现出令人印象深刻的涌现能力，但相关昂贵的 API 成本极大限制了其实际应用。以往如思维链（CoT）和思维树（ToT）等研究主要关注提升准确率，却忽视了快速增加的 API 成本，这对于具有巨大解空间的开放式现实任务尤为棘手。受人类认知双过程理论启发，我们提出“Synergy of Thoughts”（SoT），以释放不同规模混合 LLMs 的高效推理协同潜力。默认情况下，SoT 使用较小规模语言模型生成多个低成本直觉思维，这类似于系统1产生的并行直觉。随后我们设计了一个置信度评估器，对直觉思维进行交叉评估，并引入可控阈值机制来判断它们之间的相互冲突。若直觉思维出现冲突，SoT 将调用更大规模语言模型的反思推理，以模拟系统2的干预，从而覆盖直觉思维并纠正推理结果。该框架与模型无关且无需训练，可灵活使用各种现成 LLMs 实现。在六项代表性推理任务上的实验表明，SoT 将 API 成本大幅降低了 38.3%-75.1%，同时实现了最先进的推理准确率和解多样性。值得注意的是，在开放式任务上的平均 token 成本降低高达 69.1%。
英文摘要：
Large language models (LLMs) have shown impressive emergent abilities in a wide range of tasks, but the associated expensive API cost greatly limits the real application. Previous works like chain-of-thought (CoT) and tree-of-thoughts (ToT) have predominately focused on enhancing accuracy, but overlook the rapidly increasing API cost, which could be particularly problematic for open-ended real-world tasks with huge solution spaces. Motivated by the dual process theory of human cognition, we propose "Synergy of Thoughts"(SoT) to unleash the synergistic potential of hybrid LLMs with different scales for efficient reasoning. By default, SoT uses smaller-scale language models to generate multiple low-cost intuitive thoughts, which resembles the parallel intuitions produced by System 1. We then design a confidence evaluator where the intuitive thoughts are cross-evaluated and introduce a controllable threshold mechanism to decide their mutual conflict. If these intuitive thoughts exhibit conflicts, SoT will invoke the reflective reasoning of scaled-up language models to emulate the intervention of System 2, which will override the intuitive thoughts and rectify the reasoning results. This framework is model-agnostic and training-free, which can be flexibly implemented with various off-the-shelf LLMs. Experiments on six representative reasoning tasks show that SoT substantially reduces the API cost by 38.3%-75.1%, and simultaneously achieves state-of-the-art reasoning accuracy and solution diversity. Notably, the average token cost reduction on open-ended tasks reaches up to 69.1%.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
