A Survey of Intelligent Design and Manufacturing in the Era of Large Models: Research Status, Technical Challenges, and Application Prospects

doi:10.3901/JME.260645

Abstract

Abstract: In the era of large models, intelligent technology systems underpinned by generative pretraining and supported by post-training alignment and tool augmentation are rapidly penetrating industrial verticals, driving intelligent design and manufacturing from “local digitalization” toward “global cognitive intelligence.” This study systematically reviews recent progress in large-model-enabled intelligent design and manufacturing, and summarizes five major research directions: multimodal engineering semantic alignment, engineering knowledge augmentation with rule/mechanism injection, conceptual generative modeling with editable representations, integrated performance/reliability inference and evaluation, and explicit reasoning optimization with multi-agent collaborative execution. From the perspective of system-level industrial applications, five key bottlenecks are further identified: mismatch of cross-domain engineering semantic state spaces, insufficient executability of knowledge and mechanism constraints, difficulty in mapping generative configurations to manufacturing-driven long operation sequences, inadequate physical consistency in performance/reliability inference under extreme conditions, and instability in long-chain collaborative optimization and autonomous execution. Accordingly, several potential application forms are discussed, including manufacturing-constraint-embedded design synthesis, constraint-dynamic knowledge services and decision support, natural-language-driven cross-stage operation sequence transformation, physics-prior-enhanced performance/reliability prediction, and explicit-reasoning-guided collaborative optimization and autonomous resource reconfiguration. Finally, case studies on an interactive CAD generation model and an aviation manufacturing domain-specific large model are presented to reveal the pathway through which large-model technologies evolve from algorithmic capabilities to engineering system capabilities.

Key words: large model technology system, intelligent design and manufacturing, multimodal alignment, knowledge augmentation, generative modeling, performance/reliability assessment, multi-agent collaboration

CLC Number:

TP301

JIA Weiqiang, LIU Xin, PENG Xiang, WANG Xun, LIU Zhenyu, LIU Hui, ZHU Weidong, CHENG Jin, TAN Jianrong. A Survey of Intelligent Design and Manufacturing in the Era of Large Models: Research Status, Technical Challenges, and Application Prospects[J]. Journal of Mechanical Engineering, 2026, 62(15): 1-23.

Add to citation manager EndNote|Reference Manager|ProCite|BibTeX|RefWorks

URL: http://www.cjmenet.com.cn/EN/10.3901/JME.260645

http://www.cjmenet.com.cn/EN/Y2026/V62/I15/1

References

[1] 澎湃新闻. ChatGPT火爆全球，这些互联网巨头坐不住了[EB/OL]. (2023-02-07)[2026-05-17]. https://www.thepaper.cn/newsDetail_forward_21830301. The Paper. ChatGPT goes viral globally，and these Internet giants can no longer sit still[EB/OL]. (2023-02-07)[2026-05-17]. https://www.thepaper.cn/newsDetail_forward_21830301.
[2] KINGMA D P，WELLING M. Auto-encoding variational bayes[EB/OL]. (2013-12-20)[2026-05-17]. https://arxiv. org/abs/1312.6114.
[3] GOODFELLOW I J，POUGET-ABADIE J，MIRZA M，et al. Generative adversarial nets[C]//Advances in Neural Information Processing Systems. 2014，1-27.
[4] HO J，JAIN A，ABBEEL P. Denoising diffusion probabilistic models[C]//Advances in Neural Information Processing Systems，2020，33：6840-6851.
[5] MIRZA M，OSINDERO S. Conditional generative adversarial nets[EB/OL]. (2014-11-06)[2026-05-17]. https://arxiv.org/abs/1411.1784.
[6] DEVLIN J，CHANG M W，LEE K，et al. BERT：Pre-training of deep bidirectional transformers for language understanding[C]//Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics：Human Language Technologies. 2019：4171-4186.
[7] RADFORD A，NARASIMHAN K，SALIMANS T，et al. Improving language understanding by generative pre-training[R]. OpenAI，2018.
[8] PETERS M E，NEUMANN M，IYYER M，et al. Deep contextualized word representations[EB/OL]. (2018-02-15) [2026-05-17]. https://arxiv.org/abs/1802.05365.
[9] LIU Y，OTT M，GOYAL N，et al. RoBERTa：A robustly optimized BERT pretraining approach[EB/OL]. (2019-07- 26)[2026-05-17]. https://arxiv.org/abs/1907.11692.
[10] TOUVRON H，LAVRIL T，IZACARD G，et al. LLaMA：Open and efficient foundation language models[EB/OL]. (2023-02-27)[2026-05-17]. https://arxiv.org/abs/2302.13971.
[11] LEWIS P，PEREZ E，PIKTUS A，et al. Retrieval- augmented generation for knowledge-intensive NLP tasks[C]//Advances in Neural Information Processing Systems. 2020，33：9459-9474.
[12] GRAVES A，WAYNE G，DANIHELKA I. Neural Turing machines[EB/OL]. (2014-10-20)[2026-05-17]. https:// arxiv.org/abs/1410.5401.
[13] BORGEAUD S，MENSCH A，HOFFMANN J，et al. Improving language models by retrieving from trillions of tokens[C]//International Conference on Machine Learning. PMLR，2022：2206-2240.
[14] RIEGEL R，GRAY A，LUUS F，et al. Logical neural networks[EB/OL]. (2020-06-23)[2026-05-17]. https:// arxiv.org/abs/2006.13155.
[15] YAO L，MAO C，LUO Y. KG-BERT：BERT for knowledge graph completion[EB/OL]. (2019-09-06) [2026-05-17]. https://arxiv.org/abs/1909.03193.
[16] WEI J，WANG X，SCHUURMANS D，et al. Chain-of-thought prompting elicits reasoning in large language models[C]//Advances in Neural Information Processing Systems. 2022，35：24824-24837.
[17] YAO S，YU D，ZHAO J，et al. Tree of thoughts：Deliberate problem solving with large language models[C]//Advances in Neural Information Processing Systems. 2023，36：11809-11822.
[18] BESTA M，BLACH N，KUBICEK A，et al. Graph of thoughts：Solving elaborate problems with large language models[C]//Proceedings of the AAAI Conference on Artificial Intelligence. 2024，38(16)：17682-17690.
[19] HUANG W，ABBEEL P，PATHAK D，et al. Language models as zero-shot planners：Extracting actionable knowledge for embodied agents[C]//International Conference on Machine Learning. PMLR，2022：9118-9147.
[20] YE H，GUI H，ZHANG A，et al. Beyond isolation：Multi-agent synergy for improving knowledge graph construction[C]//China Conference on Knowledge Graph and Semantic Computing. Singapore：Springer Nature，2024：69-81.
[21] YAO S，ZHAO J，YU D，et al. ReAct：Synergizing reasoning and acting in language models[C]//The Eleventh International Conference on Learning Representations. 2023.
[22] SCHICK T，DWIVEDI-YU J，DESSÌ R，et al. Toolformer：Language models can teach themselves to use tools[C]//Advances in Neural Information Processing Systems. 2023，36：68539-68551.
[23] KARPAS E，ABEND O，BELINKOV Y，et al. MRKL systems：A modular，neuro-symbolic architecture that combines large language models，external knowledge sources and discrete reasoning[EB/OL]. (2022-05-01) [2026-05-17]. https://arxiv.org/abs/2205.00445.
[24] WU Q，BANSAL G，ZHANG J，et al. AutoGen：Enabling next-gen LLM applications via multi-agent conversations[C]//First Conference on Language Modeling. 2024：1-7.
[25] WANG G，XIE Y，JIANG Y，et al. Voyager：An open-ended embodied agent with large language models[EB/OL]. (2023-05-25)[2026-05-17]. https://arxiv.org/abs/2305.16291.
[26] STRATTON J. An introduction to Microsoft Copilot[M]. Berkeley：Apress，2024.
[27] RANE N，CHOUDHARY S，RANE J. Gemini versus ChatGPT：Applications，performance，architecture，capabilities，and implementation[J]. Journal of Applied Artificial Intelligence，2024，5(1)：69-93.
[28] ZHU Q，GUO D，SHAO Z，et al. DeepSeek-Coder-V2：Breaking the barrier of closed-source models in code intelligence[EB/OL]. (2024-06-17)[2026-05-17]. https:// arxiv.org/abs/2406.11931.
[29] LUTZE R，WALDHÖR K. Generating specifications from requirements documents for smart devices using large language models (LLMs)[C]//International Conference on Human-Computer Interaction. Cham：Springer Nature Switzerland，2024：94-108.
[30] ARVIDSSON S, AXELL J. Prompt engineering guidelines for LLMs in requirements engineering[R]. Lund：Lund University，2023.
[31] MA Q，PENG W，YANG C，et al. What should we engineer in prompts? Training humans in requirement-driven LLM use[J]. ACM Transactions on Computer-Human Interaction，2025，32(4)：1-27.
[32] MIRABITO Y，LIU X，GOUCHER-LAMBERT K. Characterizing design rationales using computational linguistics and human evaluations[J]. Journal of Mechanical Design，2025，147(12)：121404.
[33] YUE G，LIU J，ZHANG W. Extracting design rationale in technical literature to provide inspirational design stimuli[J]. Journal of Mechanical Design，2025，147(7)：071401.
[34] DORIS A C，GRANDI D，TOMICH R，et al. DesignQA：A multimodal benchmark for evaluating large language models' understanding of engineering documentation[J]. Journal of Computing and Information Science in Engineering，2025，25(2)：021009.
[35] WU S，KHASAHMADI A H，KATZ M，et al. CADVLM：Bridging language and vision in the generation of parametric CAD sketches[C]//European Conference on Computer Vision. Cham：Springer Nature Switzerland，2024：368-384.
[36] PICARD C，EDWARDS K M，DORIS A C，et al. From concept to manufacturing：Evaluating vision-language models for engineering design[J]. Artificial Intelligence Review，2025，58(9)：288.
[37] KARADENIZ A S，MALLIS D，MEJRI N，et al. PICASSO：A feed-forward framework for parametric inference of CAD sketches via rendering self-supervision[C]//2025 IEEE/CVF Winter Conference on Applications of Computer Vision. IEEE，2025：6475-6484.
[38] 董清钰，梅标，傅云，等. 集成改进半监督分割方法的航空发动机短舱声衬孔检测系统[J]. 光学精密工程，2024，32(23)：3457-3468. DONG Qingyu，MEI Biao，FU Yun，et al. Aero-engine nacelle acoustic liner hole detection system integrating improved semi-supervised segmentation method[J]. Optics and Precision Engineering，2024，32(23)：3457-3468.
[39] MCKINNEY M，GARLAND A，CILLESSEN D，et al. Unsupervised multimodal fusion of in-process sensor data for advanced manufacturing process monitoring[J]. Journal of Manufacturing Systems，2025，78：271-282.
[40] ZAMIELA C，JIANG Z，STOKES R，et al. Deep multi-modal U-Net fusion methodology of thermal and ultrasonic images for porosity detection in additive manufacturing[J]. Journal of Manufacturing Science and Engineering，2023，145(6)：061009.
[41] SHENG Y，ZHANG G，ZHANG Y，et al. A multimodal data sensing and feature learning-based self-adaptive hybrid approach for machining quality prediction[J]. Advanced Engineering Informatics，2024，59：102324.
[42] XIE J，GUO Y，LIU D，et al. A multimodal fusion method for soldering quality online inspection[J]. Journal of Intelligent Manufacturing，2025，36(5)：3271-3284.
[43] BIANCHINI F，CALAMO M，MARINACCI M，et al. Automating industrial quality control：A multimodal LLM and RAG framework for anomaly detection[C]//IFIP International Conference on Artificial Intelligence Applications and Innovations. Cham：Springer Nature Switzerland，2025：253-266.
[44] FANG Q，XIONG G，WANG F，et al. Large language models as few-shot defect detectors for additive manufacturing[C]//2024 China Automation Congress. IEEE，2024：6900-6905.
[45] ZHAO S, ZHANG G, LIU S, et al. Interpretable verification mechanism for trustworthy industrial large model in intelligent manufacturing[J]. Engineering，2025：1-22.
[46] ZHANG D，LIU Z，JIA W，et al. Dual attention graph convolutional network for relation extraction[J]. IEEE Transactions on Knowledge and Data Engineering，2024，36(2)：530-543.
[47] 张栋豪，刘振宇，郏维强，等. 知识图谱在智能制造领域的研究现状及其应用前景综述[J]. 机械工程学报，2021，57(5)：90-113. ZHANG Donghao，LIU Zhenyu，JIA Weiqiang，et al. Review on research status and application prospects of knowledge graph in intelligent manufacturing[J]. Journal of Mechanical Engineering，2021，57(5)：90-113.
[48] HUANG Y，YU S，CHU J，et al. Design knowledge graph-aided conceptual product design approach based on joint entity and relation extraction[J]. Journal of Intelligent & Fuzzy Systems，2023，44(3)：5333-5355.
[49] WANG R，SUN Y，PENG T，et al. A knowledge graph-aided decision guidance method for product conceptual design[J]. Journal of Engineering Design，2025，36(2)：216-255.
[50] HUANG Y，WANG R，LI Y，et al. Knowledge graph-driven methodology for complex product architecture solution generation and simulation verification[J]. Advanced Engineering Informatics，2025，68：103590.
[51] GARCIA C I，DIBATTISTA M A，LETELIER T A，et al. Framework for LLM applications in manufacturing[J]. Manufacturing Letters，2024，41：253-263.
[52] XIA L，LI C，ZHANG C，et al. Leveraging error-assisted fine-tuning large language models for manufacturing excellence[J]. Robotics and Computer-Integrated Manufacturing，2024，88：102728.
[53] KERNAN FREIRE S，WANG C，FOOSHERIAN M，et al. Knowledge sharing in manufacturing using LLM-powered tools：User study and model benchmarking[J]. Frontiers in Artificial Intelligence，2024，7：1293084.
[54] ZHOU B，LI X，LIU T，et al. CausalKGPT：Industrial structure causal knowledge-enhanced large language model for cause analysis of quality problems in aerospace product manufacturing[J]. Advanced Engineering Informatics，2024，59：102333.
[55] ÁLVARO J A H，BARREDA J G. An advanced retrieval-augmented generation system for manufacturing quality control[J]. Advanced Engineering Informatics，2025，64：103007.
[56] SHI D，LIEDL P，BAUERNHANSL T. Interoperable information modelling leveraging asset administration shell and large language model for quality control toward zero defect manufacturing[J]. Journal of Manufacturing Systems，2024，77：678-696.
[57] MA K，GRANDI D，MCCOMB C，et al. Conceptual design generation using large language models[C]//International Design Engineering Technical Conferences and Computers and Information in Engineering Conference. ASME，2023：1-7.
[58] NOMAGUCHI Y，KUROISHI K，SYAMIMI A，et al. Creating design catalog from patent documents with large language model for design concept generation[J]. Proceedings of the Design Society，2025，5：1051-1060.
[59] LI S，ZHOU X，LIU Y，et al. Agile conceptual design and validation based on multi-source product data and large language models：A review，framework，and outlook[J]. Journal of Engineering Design，2025，36(4)：473-503.
[60] JING L，ZHANG H，DOU Y，et al. Conceptual design decision-making considering multigranularity heterogeneous evaluation semantics with uncertain beliefs[J]. Expert Systems with Applications，2024，244：122963.
[61] CHEN L，ZUO H，CAI Z，et al. Toward controllable generative design：A conceptual design generation approach leveraging the function-behavior-structure ontology and large language models[J]. Journal of Mechanical Design，2024，146(12)：121401.
[62] FANG C，ZHU Y，FANG L，et al. Generative AI-enhanced human-AI collaborative conceptual design：A systematic literature review[J]. Design Studies，2025，97：101300.
[63] 程锦，叶虎强，谭建荣，等. 三维CAD技术研究进展及其发展趋势综述[J]. 机械工程学报，2023，59(23)：158-185. CHENG Jin，YE Huqiang，TAN Jianrong，et al. Review of research progress and development trends of three-dimensional CAD technology[J]. Journal of Mechanical Engineering，2023，59(23)：158-185.
[64] WU R，XIAO C，ZHENG C. DeepCAD：A deep generative network for computer-aided design models[C]//Proceedings of the IEEE/CVF International Conference on Computer Vision. 2021：6772-6782.
[65] XU X，JAYARAMAN P K，LAMBOURNE J G，et al. Hierarchical neural coding for controllable CAD model generation[EB/OL]. (2023-07-01)[2026-05-17]. https:// arxiv.org/abs/2307.00149.
[66] QI D，WANG C，XU J，et al. Pointer-CAD：Unifying B-Rep and command sequences via pointer-based edges and faces selection[EB/OL]. (2026-03-05)[2026-05-17]. https://arxiv.org/abs/2603.04337.
[67] ZHANG A，JIA W，ZOU Q，et al. Diffusion-CAD：Controllable diffusion model for generating computer-aided design models[J]. IEEE Transactions on Visualization and Computer Graphics，2025，31(12)：10188-10199.
[68] CHEN H，JOGLEKAR A，RUBINSTEIN Z，et al. Generative manufacturing：A requirements and resource- driven approach to part making[EB/OL]. (2024-09-04) [2026-05-17]. https://arxiv.org/abs/2409.03089.
[69] KUSIAK A. Generative artificial intelligence in smart manufacturing[J]. Journal of Intelligent Manufacturing，2025，36(1)：1-3.
[70] SANCHEZ-GONZALEZ A，GODWIN J，PFAFF T，et al. Learning to simulate complex physics with graph networks[C]//International Conference on Machine Learning. PMLR，2020：8459-8468.
[71] PFAFF T，FORTUNATO M，SANCHEZ-GONZALEZ A，et al. Learning mesh-based simulation with graph networks[J]. //International Conference on Learning Representations. 2021.
[72] HARTMANN S，MURUA O，ARRIZUBIETA J I，et al. Physics-informed neural network approach to speed up Laser-DED modelling[J]. Procedia CIRP，2024，124：352-357.
[73] KARNIADAKIS G E，KEVREKIDIS I G，LU L，et al. Physics-informed machine learning[J]. Nature Reviews Physics，2021，3(6)：422-440.
[74] BRANDSTETTER J，WORRALL D，WELLING M. Message passing neural PDE solvers[EB/OL]. (2022-02- 07)[2026-05-17]. https://arxiv.org/abs/2202.03376.
[75] BASTEK J H，SUN W，KOCHMANN D M. Physics-informed diffusion models[EB/OL]. (2024-03-21) [2026-05-17]. https://arxiv.org/abs/2403.14404.
[76] SHU D，LI Z，FARIMANI A B. A physics-informed diffusion model for high-fidelity flow field reconstruction[J]. Journal of Computational Physics，2023，478：111972.
[77] LIU Y，LIN Q，PAN W，et al. SR-M-GAN：A generative model for high-fidelity stress fields prediction of the composite bolted joints[J]. Advanced Engineering Informatics，2024，61：102537.
[78] LIU Y，YU W，LIN Q，et al. TF-F-GAN：A GAN-based model to predict the assembly physical fields under multi-modal variables fusion on vision transformer[J]. Advanced Engineering Informatics，2024，62：102871.
[79] CHEN Y，LIU C. Causal inference based transfer learning with LLMs：An efficient framework for industrial RUL prediction[EB/OL]. (2025-03-22)[2026-05-17]. https:// arxiv.org/abs/2503.17686.
[80] REN J, LIU X, WANG T, et al. PHM-GPT：A large language model for prognostics and health management[J/OL]. (2025-11-13)Engineering，2025，https://doi.org/10.1016/j.end.2025.11.001.
[81] JADHAV Y，PAK P，FARIMANI A B. LLM-3D print：Large language models to monitor and control 3D printing[J]. Additive Manufacturing，2025，114：105027.
[82] LI M，LOU S，ZHENG H，et al. A cognitive analysis-based key concepts derivation approach for product design[J]. Expert Systems with Applications，2024，236：121289.
[83] LIU M，HU Y. Application potential of stable diffusion in different stages of industrial design[C]//International Conference on Human-Computer Interaction. Cham：Springer Nature Switzerland，2023：590-609.
[84] CHANG C，WANG S，ZHANG J，et al. LLMScenario：Large language model driven scenario generation[J]. IEEE Transactions on Systems，Man，and Cybernetics：Systems，2024，54(11)：6581-6594.
[85] ZHANG J，WANG W，GUO S，et al. Solving general natural-language-description optimization problems with large language models[EB/OL]. (2024-07-10)[2026-05- 17]. https://arxiv.org/abs/2407.07924.
[86] ZHANG S，LIU S，LU N，et al. LLM-driven instance-specific heuristic generation and selection[EB/OL]. (2025-05-31)[2026-05-17]. https:// arxiv.org/abs/2506.00490.
[87] ZHANG R，LIU F，LIN X，et al. Understanding the importance of evolutionary search in automated heuristic design with large language models[C]//International Conference on Parallel Problem Solving from Nature. Cham：Springer Nature Switzerland，2024：185-202.
[88] 王沐晨，朱泓宇，宋学官. 大语言模型赋能优化设计初探[J/OL]. 机械工程学报，1-15[2026-05-17]. https://link. cnki.net/urlid/11.2187.TH.20251119.1411.036. WANG Muchen，ZHU Hongyu，SONG Xueguan. Preliminary study on optimization design empowered by large language models[J/OL]. Journal of Mechanical Engineering，1-15[2026-05-17]. https://link.cnki. net/urlid/11.2187.TH.20251119.1411.036.
[89] ZHANG Z，ZHAO H，ZHAO W，et al. LLM-assisted non-dominated sorting genetic algorithm for solving distributed heterogeneous no-wait permutation flowshop scheduling[J]. Applied Sciences，2025，15(18)：10131.
[90] LI R，WANG L，SANG H，et al. LLM-assisted automatic memetic algorithm for lot-streaming hybrid job shop scheduling with variable sublots[J]. IEEE Transactions on Evolutionary Computation，2025：1-28.
[91] ZHOU Q，LI M，LI W，et al. LLM-assisted advanced planning and scheduling framework for smart manufacturing systems[C]//Proceedings of the 202517th International Conference on Computer Modeling and Simulation. 2025：162-167.
[92] ÇETINKAYA İ O，BÜYÜKTAHTAKIN İ E，SHOJAEE P，et al. Discovering heuristics with large language models (LLMs) for mixed-integer programs：Single-machine scheduling[J]. Computers & Operations Research，2025：107325.
[93] WANG Y，WANG J，CHU Z. Multi-agent large language models as evolutionary optimizers for scheduling optimization[J]. Computers & Industrial Engineering，2025，206：111197.
[94] LIU Y，ZHANG H，LI Z，et al. Optimizing the utilization of large language models via schedule optimization：An exploratory study[C]//Proceedings of the 18th ACM/IEEE International Symposium on Empirical Software Engineering and Measurement. 2024：84-95.
[95] XIA Y，SHENOY M，JAZDI N，et al. Towards autonomous system：Flexible modular production system enhanced with large language model agents[C]//2023 IEEE 28th International Conference on Emerging Technologies and Factory Automation. IEEE，2023：1-8.
[96] LIU Z，ZENG R，WANG D，et al. Agents4PLC：Automating closed-loop PLC code generation and verification in industrial control systems using LLM-based agents[EB/OL]. (2024-10-18)[2026-05-17]. https://arxiv.org/abs/2410.14209.
[97] FAN J，ZHENG P. A vision-language-guided robotic action planning approach for ambiguity mitigation in human-robot collaborative manufacturing[J]. Journal of Manufacturing Systems，2024，74：1009-1018.
[98] GKOURNELOS C，KONSTANTINOU C，ANGELAKIS P，et al. Enabling seamless human-robot collaboration in manufacturing using LLMs[C]//European Symposium on Artificial Intelligence in Manufacturing. Cham：Springer Nature Switzerland，2023：81-89.
[99] WANG Z，ZHOU X，MAO J，et al. LLM-MTMP：A large language model-based multi-agent task and motion planning framework for power inspection robots[J]. Journal of Industrial Information Integration，2025：101014.
[100] BARBOSA R，SANTOS R，NOVAIS P. Collaborative problem-solving with LLM：A multi-agent system approach to solve complex tasks using AutoGen[C]// International Conference on Practical Applications of Agents and Multi-Agent Systems. Cham：Springer Nature Switzerland，2024：203-214.
[101] LI L，CHEN Z，KAN Z. Environment-driven and LLM-guided multi-robot task inference and allocation under temporal logic specifications[J]. IEEE Transactions on Automation Science and Engineering，2026，23：2925-2940.
[102] WASEEM M，BHATTA K，LI C，et al. Pretrained LLMs as real-time controllers for robot operated serial production line[EB/OL]. (2025-03-05)[2026-05-17]. https:// arxiv.org/abs/2503.03889.
[103] LING S，WANG Y，FAN C，et al. ELHPlan：Efficient long-horizon task planning for multi-agent collaboration[EB/OL]. (2025-09-30)[2026-05-17]. https:// arxiv.org/abs/2509.24230.