发表机构
Shanghai JiaoTong University（上海交通大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
该研究针对大型语言模型逻辑推理的缺陷，提出结合逻辑知识图谱与动态求解器路由的神经-符号架构，在基准任务上优于现有方法，实现了可验证的高准确率逻辑推理。
AI 中文摘要
大型语言模型（LLMs）在自然语言理解方面展现出卓越能力，但在严格的多步推理任务中表现不佳，常出现幻觉与不一致问题。现有解决方案如思维链（CoT）缺乏严格的验证机制，而标准检索增强生成（RAG）往往会遗漏逻辑任务固有的复杂结构依赖。为弥合这一差距，本文提出一种神经-符号架构，将逻辑知识图谱（LKG）与动态求解器路由相结合。具体而言，我们引入基于本体的LKG，将逻辑规则与约束视为一级拓扑节点，实现对从文本中提取的依赖关系的显式建模；进一步设计逻辑路由器，通过拓扑感知混合检索机制支持，将任务动态分配至最优符号引擎。在逻辑推理基准上的实验结果表明，我们的框架显著优于当前最优的提示工程与RAG基线，实现了更高的准确率与可验证的推理路径。
英文摘要
Large Language Models (LLMs) have demonstrated remarkable proficiency in natural language understanding, yet they struggle with strict multi-step reasoning, frequently suffering from hallucinations and inconsistency. Existing solutions like Chain-of-Thought (CoT) lack rigorous verification mechanisms, while standard Retrieval-Augmented Generation (RAG) often misses the complex, structural dependencies inherent in logical tasks. To bridge this gap, we propose a Neuro-Symbolic architecture that integrates a Logical Knowledge Graph (LKG) with dynamic solver routing. Specifically, we introduce an ontology-based LKG that treats logical rules and constraints as first-class topological nodes, enabling explicit modeling of dependencies extracted from text. We further design a Logic Router to dynamically dispatch tasks to the optimal symbolic engine, which is supported by a topology-aware hybrid retrieval mechanism. Experimental results on logical reasoning benchmarks demonstrate that our framework significantly outperforms state-of-the-art prompting and RAG baselines, delivering higher accuracy and verifiable reasoning paths.
Comments19 pages, 9 figures
