mlsanityData to 7 Oct 2026

PathWise: Planning through World Model for Automated Heuristic Design via Self-Evolving LLMs

arXiv 2601.20539 ↗First posted 28 Jan 2026cs.AI cs.CLcited by 5

On the map: World models → World models for language agents

also in 1 more
Abstract (arXiv)

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.

Key facts. Cited by 5 papers in the map (data through 7 Oct 2026). Placed in World models for language agents, one of 11 directions in World models. Builds on 64 of its 116 references that are in the map. First posted on arXiv 28 Jan 2026.

Builds on

64 of 116 references.

Closest ancestors · 3

Same direction, newest first

  1. 2024Deliberate Reasoning in Language Models as Structure-Aware Planning with an Accurate World Model4 Oct
  2. 2023Reasoning with Language Model is Planning with World Model24 May
  3. 2023Large Language Models as Commonsense Knowledge for Large-Scale Task Planning23 May

Other references · 61

  1. 2022Chain-of-Thought Prompting Elicits Reasoning in Large Language Modelscited by 12,302
  2. 2021Evaluating Large Language Models Trained on Codecited by 8,053
  3. 2023Reflexion: Language Agents with Verbal Reinforcement Learningcited by 4,064
  4. 2023Self-Refine: Iterative Refinement with Self-Feedbackcited by 3,317
  5. 2022Least-to-Most Prompting Enables Complex Reasoning in Large Language Modelscited by 1,397
  6. 2023WizardLM: Empowering large pre-trained language models to follow complex instructionscited by 1,116
  7. 2022Large Language Models Are Human-Level Prompt Engineerscited by 1,020
  8. 2025A-MEM: Agentic Memory for LLM Agentscited by 1,000
  9. 2025OpenAI GPT-5 System Cardcited by 910
  10. 2023Large Language Models as Optimizerscited by 830
  11. 2025DeepSeek-V3.2: Pushing the Frontier of Open Large Language Modelscited by 755
  12. 2025Mem0: Building Production-Ready AI Agents with Scalable Long-Term Memorycited by 722
38 more
  1. 2023Automatic Prompt Optimization with "Gradient Descent" and Beam Searchcited by 628
  2. 2023Promptbreeder: Self-Referential Self-Improvement Via Prompt Evolutioncited by 521
  3. 2025Agentic Context Engineering: Evolving Contexts for Self-Improving Language Modelscited by 349
  4. 2020POMO: Policy Optimization with Multiple Optima for Reinforcement Learningcited by 341
  5. 2024ReEvo: Large Language Models as Hyper-Heuristics with Reflective Evolutioncited by 266
  6. 2024Evolution of Heuristics: Towards Efficient Automatic Algorithm Design Using Large Language Modelcited by 258
  7. 2025ReasoningBank: Scaling Agent Self-Evolving with Reasoning Memorycited by 253
  8. 2022Making Large Language Models Better Reasoners with Step-Aware Verifiercited by 231
  9. 2025Reasoning Models Can Be Effective Without Thinkingcited by 182
  10. 2021Monte Carlo Tree Search: A Review of Recent Modifications and Applicationscited by 158
  11. 2021Explaining Answers with Entailment Treescited by 148
  12. 2022Evolution through Large Modelscited by 131
  13. 2023EvoPrompting: Language Models for Code-Level Neural Architecture Searchcited by 129
  14. 2023Large Language Models as Evolutionary Optimizerscited by 122
  15. 2023Language Model Crossover: Variation through Few-Shot Promptingcited by 119
  16. 2021Learning to Iteratively Solve Routing Problems with Dual-Aspect Collaborative Transformercited by 108
  17. 2025Monte Carlo Tree Search for Comprehensive Exploration in LLM-Based Automatic Heuristic Designcited by 100
  18. 2024Evolutionary Computation in the Era of Large Language Model: Survey and Roadmapcited by 89
  19. 2023LLMatic: Neural Architecture Search via Large Language Models and Quality Diversity Optimizationcited by 86
  20. 2023Learning to Search Feasible and Infeasible Regions of Routing Problems with Flexible Neural k-Optcited by 81
  21. 2023DeepACO: Neural-enhanced Ant Systems for Combinatorial Optimizationcited by 77
  22. 2023Split and Merge: Aligning Position Biases in LLM-based Evaluatorscited by 71
  23. 2024HSEvo: Elevating Automatic Heuristic Design with Diversity-Driven Harmony Search and Genetic Algorithm Using LLMscited by 68
  24. 2023Harnessing the Power of Large Language Models for Natural Language to First-Order Logic Translationcited by 59
  25. 2024Large Language Models As Evolution Strategiescited by 57
  26. 2024Ant Colony Sampling with GFlowNets for Combinatorial Optimizationcited by 53
  27. 2023Algorithm Evolution Using Large Language Modelcited by 52
  28. 2024Self-evolving Agents with reflective and memory-augmented abilitiescited by 52
  29. 2024Multi-objective Evolution of Heuristic Using Large Language Modelcited by 52
  30. 2025Sample, Scrutinize and Scale: Effective Inference-Time Search by Scaling Verificationcited by 49
  31. 2025Diverse Preference Optimizationcited by 48
  32. 2024Evolving Code with A Large Language Modelcited by 42
  33. 2025Self-Supervised Prompt Optimizationcited by 37
  34. 2025EoH-S: Evolution of Heuristic Set using LLMs for Automated Heuristic Designcited by 32
  35. 2023Bounce: Reliable High-Dimensional Bayesian Optimization for Combinatorial and Mixed Spacescited by 27
  36. 2025Diverse Preference Learning for Capabilities and Alignmentcited by 25
  37. 2025On the Effect of Sampling Diversity in Scaling LLM Inferencecited by 18
  38. 2025Curiosity-Driven Reinforcement Learning from Human Feedbackcited by 14

50 most cited of 61 All 61 (JSON)

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2026: 5.

5 papers.

  1. 2026Adaptive Information Control for Search-Augmented LLM Reasoningcited by 8
  2. 2026AHD Agent: Agentic Reinforcement Learning for Automatic Heuristic Designcited by 3
  3. 2026DIVE: Unlocking Self-Improvement in Frozen Language Models Through Diversity-Driven Skill Evolutioncited by 2
  4. 2026MOSAIC: Adversarial Co-evolution of Specialist Heuristics and Problem Instances for LLM-based Automated Heuristic Designcited by 0
  5. 2026FormuEvo: LLM-Guided Evolution for Discovering Solver-Efficient Mixed-Integer Programming Formulationscited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2026OR-Agent: Bridging Evolutionary Search and Structured Research for Automated Heuristic Design
  2. 2025HeuriGym: An Agentic Benchmark for LLM-Crafted Heuristics in Combinatorial Optimization
  3. 2026SimpleEvol: An Agent-Loop Framework for LLM-Driven Automated Heuristic Design with Minimal Human Priors
  4. 2025Complex LLM Planning via Automated Heuristics Discovery
  5. 2024Query-Efficient Planning with Language Models
  6. 2026HMACE: Heterogeneous Multi-Agent Collaborative Evolution for Combinatorial Optimization
  7. 2026LLM-Evolved Domain-Independent Heuristics for Symbolic AI Planning
  8. 2025Large Language Models for Planning: A Comprehensive and Systematic Survey