Large Language Models (LLMs) have enabled automated heuristic design (AHD) for combinatorial optimization problems (COPs), but existing frameworks' reliance on fixed evolutionary rules and static prompt templates often leads to myopic heuristic generation, redundant evaluations, and limited reasoning about how new heuristics should be derived. We propose a novel multi-agent reasoning framework, referred to as Planning through World Model for Automated Heuristic Design via Self-Evolving LLMs (PathWise), which formulates heuristic generation as a sequential decision process over an entailment graph serving as a compact, stateful memory of the search trajectory. This approach allows the system to carry forward past decisions and reuse or avoid derivation information across generations. A policy agent plans evolutionary actions, a world model agent generates heuristic rollouts conditioned on those actions, and critic agents provide routed reflections summarizing lessons from prior steps, shifting LLM-based AHD from trial-and-error evolution toward state-aware planning through reasoning. Experiments across diverse COPs show that PathWise converges faster to better heuristics, generalizes across different LLM backbones, and scales to larger problem sizes.

翻译：大语言模型（LLMs）已能实现组合优化问题（COPs）的自动启发式设计（AHD），但现有框架依赖固定的进化规则和静态提示模板，常导致短视启发式生成、冗余评估以及对新启发式推导逻辑的有限推理。我们提出一种新型多智能体推理框架，即通过世界模型规划实现自演化大语言模型的自动启发式设计（PathWise），该方法将启发式生成形式化为在蕴含图上的序贯决策过程——该图作为搜索轨迹的紧凑状态化记忆。该机制使系统能延续过往决策，并在不同生成期之间复用或规避推导信息。策略智能体规划进化动作，世界模型智能体基于动作生成启发式展开，评论智能体则提供路由式反思，总结先前步骤经验。这将基于LLM的AHD从试错进化转变为通过推理进行状态感知规划。跨多种COPs的实验表明，PathWise能更快收敛至更优启发式，在不同LLM骨干模型上具备泛化能力，并可扩展至更大规模问题。