面向资源受限边缘设备的神经符号路由可靠推理
原标题：Neurosymbolic Routing for Reliable Reasoning on Resource-Constrained Edge Devices
AI 摘要
该论文提出一种神经符号路由器，将查询分类为算术、代数、形式逻辑和文字题四类，前三类交由确定性求解器处理，仅文字题使用小语言模型。路由逻辑通过L*语法推断算法学习DFA实现，以SLM作为成员查询预言机、标注数据作为等价预言机。在Raspberry Pi 4B上运行Phi-4-mini（3.8B，Q6_K量化），对DeepMind Mathematics、GSM8K和RuleTaker的100个未测试提示进行评估，路由准确率达100%，整体准确率98.3%，优于最强基线Program-of-Thought的72.0%和工具调用智能体的58.7%，且延迟降低8.8倍、能耗降低2.8倍。
正文节选
Neurosymbolic Routing for Reliable Reasoning on Resource-Constrained Edge Devices Abstract Running a language model on edge hardware provides private and low-latency reasoning without a network connection, and yet the small models that fit on such devices are unreliable on the tasks computers are expected to handle well, such as arithmetic, algebra, and formal logic problems. We argue that much of this unreliability is avoidable. Many queries appearing to demand reasoning are in fact structurall