计算机科学

计算机科学 ›› 2026, Vol. 53 ›› Issue (9): 101-109.doi: 10.11896/jsjkx.260100163

• 大语言模型技术研究及应用 • 上一篇    下一篇

基于大模型的行动方案生成思维链模型构建研究

雷坤翰1,3, 陶九阳2, 李辰宇1, 钟琪1, 叶林发2   

  1. 1 国防大学研究生院 北京 100081
    2 国防大学联合作战学院 北京 100081
    3 中国人民解放军65426部队 山东 潍坊 261021
  • 收稿日期:2026-01-26 修回日期:2026-04-29 出版日期:2026-09-15 发布日期:2026-09-10
  • 通讯作者: 陶九阳(taojiuyang@126.com)
  • 作者简介:(594908179@qq.com)

Method for Generating Operational Action Plans Based on Large Language Model Guided byChain-of-Thought

LEI Kunhan1,3, TAO Jiuyang2, LI Chenyu1, ZHONG Qi1, YE Linfa2   

  1. 1 Graduate School,National Defense University,Beijing 100081,China
    2 Joint Operations College,National Defense University,Beijing 100081,China
    3 Unit 65426,Chinese People’s Liberation Army,Weifang,Shandong 261021,China
  • Received:2026-01-26 Revised:2026-04-29 Published:2026-09-15 Online:2026-09-10
  • About author:LEI Kunhan,born in 1993,postgra-duate.His main research interests include intelligent gaming and adversarial technology.
    TAO Jiuyang,born in 1983,Ph.D,senior engineer.His main research interest is intelligent gaming.

PDF (PC)

318

摘要： 思维链是一种提示工程技术,其内涵是通过设计简单提示或者结合领域知识精心设计推理逻辑来引导大模型生成中间推理步骤,从而更好地解决复杂问题。目前思维链已在复杂推理任务中展现出显著的成效。在指挥决策、作战筹划等高度精细化的军事领域,需要精心设计思维链来模拟人类指挥员筹划生成作战方案的步骤和推理逻辑。通过针对判断上级意图、分析战场环境、开展作战筹划等内容描述指挥员思维过程,构建思维图景模型。根据指挥员思维图景模型,构建了一个作战方案生成和优化的双向思维链模型,创造性思维链负责生成作战方案,批判性思维链负责对生成的方案进行威胁、风险分析和优化调整。所构建的模型为后续用思维链引导大模型生成作战行动方案奠定了理论基础。

关键词: 大模型, 思维链, 作战筹划, 作战行动, 指挥员思维图景

Abstract: Chain-of-Thought(CoT) is a prompt engineering technique whose core connotation lies in guiding large language mo-dels to generate intermediate reasoning steps by designing simple prompts or elaborately constructing reasoning logic combined with domain knowledge,thereby better solving complex problems.Currently,CoT has demonstrated remarkable effectiveness in complex reasoning tasks.In highly refined military domains such as command decision-making and operational planning,it is necessary to carefully design CoT to simulate the procedural steps and reasoning logic of human commanders in formulating opera-tional plans.This study describes the commander’s thinking process,including judging superior intentions,analyzing the battlefield environment,and conducting operational planning,and constructs a commander’s thinking map model.Based on this model,a two-way CoT model for operational plan generation and optimization is established:the creative CoT is responsible for generating operational plans,while the critical CoT is dedicated to threat and risk analysis,optimization,and adjustment of the generated plans.The models constructed in this research lay a theoretical foundation for subsequent studies on guiding large language models to generate operational action plans using CoT.

Key words: Large language model, Chain-of-Thought, Operational planning, Combat operations, Commander’s thinking map

中图分类号: 

  • E919
[1] WEI J,WANG X,SCHUURMANS D,et al.Chain-of-thoughtprompting elicits reasoning in large language models [J].Advances in Neural Information Processing Systems,2022,35:24824-24837.
[2] KOJIMA T,GU S S,REID M,et al.Large language models are zero-shot reasoners [J].Advances in Neural Information Processing Systems,2022,35:22199-22213.
[3] ZHOU Y Z,LUO J R,GU X Q,et al.A Survey of Intelligent Planning Methods from the Perspective of Large Language Models [J].Journal of System Simulation,2025,37(4):823-844.
[4] WANG L,XU W,LAN Y,et al.Plan-and-solve prompting:Improving zero-shot chain-of-thought reasoning by large language models [J].arXiv:2305.04091,2023.
[5] QIN L,CHEN Q,WEI F,et al.Cross-lingual prompting:Improving zero-shot chain-of-thought reasoning across languages [J].arXiv:2310.14799,2023.
[6] ZHAO X,LI M,LU W,et al.Enhancing zero-shot chain-of-thought reasoning in large language models through logic [J].arXiv:2309.13339,2023.
[7] VERMA P,LA N,FAVIER A,et al.Teaching LLMs to Plan:Logical Chain-of-Thought Instruction Tuning for Symbolic Planning [J].arXiv:2509.13351,2025.
[8] MITADERA T,UTSURO T.CoT based Few-Shot Learning ofNegative Comments Filtering [C] //2024 IEEE International Conference on Big Data(BigData).IEEE,2024:2989-2998.
[9] KUMAR S,MENDKE S,RAHMAN K L A,et al.Enhancing zero-shot chain of thought prompting via uncertainty-guided strategy selection [J].arXiv:2412.00353,2024.
[10] KIM S,JOO S J,KIM D,et al.The cot collection:Improving zero-shot and few-shot learning of language models via chain-of-thought fine-tuning [J].arXiv:2305.14045,2023.
[11] ZHANG Z,ZHANG A,LI M,et al.Automatic chain of thought prompting in large language models [J].arXiv:2210.03493,2022.
[12] ZHANG Y,CHEN Q,LI M,et al.AutoCAP:Towards automa-tic cross-lingual alignment planning for zero-shot chain-of-thought [J].arXiv:2406.13940,2024.
[13] XIANG K,LIU Z,JIANG Z,et al.Can Atomic Step Decomposition Enhance the Self-structured Reasoning of Multimodal Large Models?[J].arXiv:2503.06252,2025.
[14] DENG Y,PRASAD K,FERNANDEZ R,et al.Implicit chain of thought reasoning via knowledge distillation [J].arXiv:2311.01460,2023.
[15] WEI X,LIU X,ZANG Y,et al.SIM-CoT:Supervised ImplicitChain-of-Thought [J].arXiv:2509.20317,2025.
[16] DENG Y,CHOI Y,SHIEBER S.From explicit cot to implicit cot:Learning to internalize cot step by step [J].arXiv:2405.14838,2024.
[17] ZHENG T,CHEN L,HAN S,et al.Learning to Reason viaMixture-of-Thought for Logical Reasoning [J].arXiv:2505.15817,2025.
[18] ZHENG G,YANG B,TANG J,et al.Ddcot:Duty-distinct chain-of-thought prompting for multimodal reasoning in language models [J].Advances in Neural Information Processing Systems,2023,36:5168-5191.
[19] MONDAL D,MODI S,PANDA S,et al.Kam-cot:Knowledgeaugmented multimodal chain-of-thoughts reasoning [C] //Proceedings of the AAAI Conference on Artificial Intelligence.2024:18798-18806.
[20] CHEN Q,QIN L,ZHANG J,et al.M3CoT:A Novel Benchmark for Multi-Domain Multi-step Multi-modal Chain-of-Thought [J].arXiv:2405.16473,2024.
[21] XU Y B,DENG S,LI Y R,et al.Research on Fine-tuning Me-thod of Classification and Grading Large Model Enhanced by Chain-of-Thought [J].Communications Technology,2025,58(9):1004-1012.
[22] CHIA Y K,CHEN G,TUAN L A,et al.Contrastive chain-of-thought prompting [J].arXiv:2311.09277,2023.
[23] WU J,YU T,CHEN X,et al.Decot:Debiasing chain-of-thought for knowledge-intensive tasks in large language models via cau-sal intervention [C] //Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics.2024:14073-14087.
[24] PAUL R.Critical Thinking [M].Beijing:China Machine Press,2025:1-2.
Viewed
Full text


Abstract

Cited
  Shared   
  Discussed   
No Suggested Reading articles found!
渝ICP备16013121号-4
地址：重庆市两江新区洪湖西路18号    邮编：401121  
电话：023-63500828（编辑部） 023-67039612（会议/专题） 023-67039623（财务/发票）
E-mail：jsjkx12@163.com
本系统由北京玛格泰克科技发展有限公司设计开发