思维定理：一种用于语言模型中溯因、演绎和归纳推理的多智能体框架
Theorem-of-Thought: A Multi-Agent Framework for Abductive, Deductive, and Inductive Reasoning in Language Models
AI总结：
针对大语言模型推理鲁棒性差、缺乏逻辑结构校验的问题，提出模拟三类推理模式的多智能体框架ToTh，通过贝叶斯信念传播评估推理一致性，在多个基准上优于现有提示技术，可生成可解释的逻辑推理链。
摘要：
大语言模型（LLM）在各类自然语言推理任务中展现出优异性能，但其推理过程仍存在鲁棒性差、难以解释的问题。思维链（CoT）等提示技术通过引导生成中间推理步骤或聚合多个输出提升了可靠性，但缺乏强制逻辑结构和评估内部一致性的机制。我们提出了思维定理（ToTh）这一全新框架，将推理建模为三个并行智能体的协作过程，每个智能体分别模拟一种不同的推理模式：溯因推理、演绎推理和归纳推理。每个智能体生成一条推理轨迹，这些轨迹被构建为形式化推理图。为评估一致性，我们采用由自然语言推理（NLI）引导的贝叶斯信念传播，为每个步骤分配置信度分数，最终选择一致性最高的图来推导答案。在符号推理基准WebOfLies和数值推理基准MultiArith上的实验表明，ToTh在多个LLM上的表现始终优于CoT、自一致性（Self-Consistency）和CoT-Decoding，同时能生成可解释、有逻辑依据的推理链。我们的发现为构建更鲁棒、更具认知启发的LLM推理方法提供了有前景的方向，代码实现可在https://github.com/KurbanIntelligenceLab/theorem-of-thought获取。
英文摘要：
Large language models (LLMs) have shown strong performance across natural language reasoning tasks, yet their reasoning processes remain brittle and difficult to interpret. Prompting techniques like Chain-of-Thought (CoT) enhance reliability by eliciting intermediate reasoning steps or aggregating multiple outputs. However, they lack mechanisms for enforcing logical structure and assessing internal coherence. We introduce Theorem-of-Thought (ToTh), a novel framework that models reasoning as collaboration among three parallel agents, each simulating a distinct mode of inference: abductive, deductive, and inductive. Each agent produces a reasoning trace, which is structured into a formal reasoning graph. To evaluate consistency, we apply Bayesian belief propagation guided by natural language inference (NLI), assigning confidence scores to each step. The most coherent graph is selected to derive the final answer. Experiments on symbolic (WebOfLies) and numerical (MultiArith) reasoning benchmarks show that ToTh consistently outperforms CoT, Self-Consistency, and CoT-Decoding across multiple LLMs, while producing interpretable and logically grounded reasoning chains. Our findings suggest a promising direction for building more robust and cognitively inspired LLM reasoning. The implementation is available at https://github.com/KurbanIntelligenceLab/theorem-of-thought.
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。
