Chain of Simulation: A Dual-Mode Reasoning Framework for Large Language Models with Dynamic Problem Routing

2026-02-02

AI Summary

Large language models struggle to dynamically select optimal reasoning strategies for diverse tasks, limiting their performance. To address this, this work proposes the Chain-of-Simulation (CoS) framework, which introduces the first dynamic dual-mode reasoning router conditioned on problem type: it employs self-consistent computational streams for mathematical problems, JSON-based symbolic state tracking for spatial reasoning, and hybrid fact extraction for multi-hop reasoning. Notably, CoS requires no additional training and achieves accuracies of 71.5%, 90.0%, and 19.0% on GSM8K, StrategyQA, and bAbI, respectively—yielding relative improvements of 1.0%, 2.5%, and 65.2% over the strongest baselines—while simultaneously reducing computational cost by 46%.

Technology Category

• Knowledge Representation and Reasoning: Common-Sense Reasoning
• Search and Optimization: Sampling/Simulation-based Search
• Cognitive Modeling & Cognitive Systems: Conceptual Inference and Reasoning

Application Category

• Search and Retrieval-Augmented AI: Large language models for search
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• User Modeling, Personalization and Recommendation: User modeling and simulation for interactive and conversational systems

Abstract

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.

Problem

• reasoning strategies
• problem routing
• large language models
• dynamic reasoning
• multi-hop inference

Innovation

• Chain of Simulation
• Dynamic Problem Routing
• Dual-Mode Reasoning
• Symbolic State Tracking
• Self-Consistency