我们提出了Chain of Simulation(CoS),一种新颖的双模式推理框架,能够动态地将问题路由到大语言模型(LLMs)中的专门推理策略。与现有的统一提示方法不同,CoS采用三种不同的推理模式:(1)用于数学问题的计算流与自一致性,(2)用于空间推理的符号状态跟踪与JSON表示,(3)用于多跳推理的混合事实提取。通过在GSM8K、StrategyQA和bAbI基准测试中使用四种最先进的模型(Gemma-3 27B、LLaMA-3.1 8B、Mistral 7B和Qwen-2.5 14B)进行全面评估,我们证明CoS在GSM8K上达到71.5%的准确率(绝对提升1.0%),在StrategyQA上达到90.0%(提升2.5%),在bAbI上达到19.0%(相对提升65.2%),相比最强基线。分析表明,问题特定的模式选择至关重要,当正确应用于数学问题时,计算模式达到81.2%的准确率,而错误路由导致0%的准确率。我们提供了详细的算法用于模式选择、状态跟踪和答案提取,确立了CoS作为提高LLM推理能力的有效方法,无需额外训练。该框架在准确性和效率之间提供了优于Self-Consistency的优越折衷,以54%更低的计算成本达到可比性能。
英文摘要:
We present Chain of Simulation (CoS), a novel dual-mode reasoning framework that dynamically routes problems to specialized reasoning strategies in Large Language Models (LLMs). Unlike existing uniform prompting approaches, CoS employs three distinct reasoning modes: (1) computational flow with self-consistency for mathematical problems, (2) symbolic state tracking with JSON representations for spatial reasoning, and (3) hybrid fact-extraction for multi-hop inference. Through comprehensive evaluation on GSM8K, StrategyQA, and bAbI benchmarks using four state-of-the-art models (Gemma-3 27B, LLaMA-3.1 8B, Mistral 7B, and Qwen-2.5 14B), we demonstrate that CoS achieves 71.5% accuracy on GSM8K (1.0% absolute improvement), 90.0% on StrategyQA (2.5% improvement), and 19.0% on bAbI (65.2% relative improvement) compared to the strongest baselines. The analysis reveals that problem-specific mode selection is crucial, with computational mode achieving 81.2% accuracy when correctly applied to mathematical problems, while misrouting leads to 0% accuracy. We provide detailed algorithms for mode selection, state tracking, and answer extraction, establishing CoS as an effective approach for improving LLM reasoning without additional training. The framework provides superior trade-offs between accuracy and efficiency compared to Self-Consistency, achieving comparable performance at 54% lower computational cost.
PiERN: Token-Level Routing for Integrating High-Precision Computation and Reasoning
Peking University(北京大学);Peking University Changsha Institute for Computing and Digital Economy(北京大学长沙计算与数字经济研究院);Beihang University(北京航空航天大学);Tsinghua University(清华大学)
Plan before Solving: Problem-Aware Strategy Routing for Mathematical Reasoning with LLMs
School of Computer Science and Technology, Xi’an Jiaotong University(西安交通大学计算机科学与技术学院);MOE KLINNS Lab, Xi’an Jiaotong University(西安交通大学MOE KLINNS实验室);Shaanxi Province Key Laboratory of Big Data Knowledge Engineering(陕西省大数据知识工程重点实验室);Sydney AI Centre, The University of Sydney(悉尼大学悉尼人工智能中心)
RouteGoT: Node-Adaptive Routing for Cost-Efficient Graph of Thoughts Reasoning
School of New Media and Communication, Tianjin University(新媒体与传播学院,天津大学);School of Computer Science and Engineering, Beihang University(计算机科学与工程学院,北航);Shanghai Jiao Tong University(上海交通大学)
ARMOR-MAD: Adaptive Routing for Heterogeneous Multi-Agent Debate in Large Language Model Reasoning
School of Cyber Science and Technology, University of Science and Technology of China(中国科学技术大学网络空间安全学院);School of Artificial Intelligence, Shenzhen Technology University(深圳技术大学人工智能学院)
RouterDC: Query-Based Router by Dual Contrastive Learning for Assembling Large Language Models
Southern University of Science and Technology(南方科技大学);The Hong Kong University of Science and Technology (Guangzhou)(香港科技大学(广州));The Hong Kong University of Science and Technology(香港科技大学)
CacheRouter: A Dual-Path Tool Routing Architecture with Cache-Preserving Main-Model Isolation for Long-Tail Tool Discovery
School of Mathematical Sciences, Beijing University of Posts and Telecommunications(北京邮电大学数学科学学院);School of Mathematics and Statistics, Chongqing University(重庆大学数学与统计学院);School of Science, Beijing Forestry University(北京林业大学理学院)
ODAR: Principled Adaptive Routing for LLM Reasoning via Active Inference
Nanyang Technological University(南洋理工大学);Carnegie Mellon University(卡内基梅隆大学);Northeast University (Qinhuangdao Campus)(东北大学(秦皇岛校区));Beihang University(北航);University of the Chinese Academy of Sciences(中国科学院大学);Renmin University of China(中国人民大学);Sun Yat-sen University(中山大学)
AI Lab, Princeton University(普林斯顿大学人工智能实验室);University of Michigan(密歇根大学);Institute for Interdisciplinary Information Sciences (IIIS), Tsinghua University(清华大学交叉信息学院);Shanghai Jiao Tong University(上海交通大学);University of Edinburgh(爱丁堡大学);King’s College London(伦敦国王学院)