基础模型与智能决策：进展、挑战与展望

Suppr 超能文献

核心技术专利：CN118964589B侵权必究

Suppr 超能文献

核心技术专利：CN118964589B侵权必究

Huang Jincai, Xu Yongjun, Wang Qi, Wang Qi Cheems, Liang Xingxing, Wang Fei, Zhang Zhao, Wei Wei, Zhang Boxuan, Huang Libo, Chang Jingru, Ma Liantao, Ma Ting, Liang Yuxuan, Zhang Jie, Guo Jian, Jiang Xuhui, Fan Xinxin, An Zhulin, Li Tingting, Li Xuefei, Shao Zezhi, Qian Tangwen, Sun Tao, Diao Boyu, Yang Chuanguang, Yu Chenqing, Wu Yiqing, Li Mengxian, Zhang Haifeng, Zeng Yongcheng, Zhang Zhicheng, Zhu Zhengqiu, Lv Yiqin, Li Aming, Chen Xu, An Bo, Xiao Wei, Bai Chenguang, Mao Yuxing, Yin Zhigang, Gui Sheng, Su Wentao, Zhu Yinghao, Gao Junyi, He Xinyu, Li Yizhou, Jin Guangyin, Ao Xiang, Zhai Xuehao, Tan Haoran, Yun Lijun, Shi Hongquan, Li Jun, Fan Changjun, Huang Kuihua, Harrison Ewen, Leung Victor C M, Qiu Sihang, Dong Yanjie, Zheng Xiaolong, Wang Gang, Zheng Yu, Wang Yuanzhuo, Guo Jiafeng, Wang Lizhe, Cheng Xueqi, Wang Yaonan, Yang Shanlin, Fu Mengyin, Fei Aiguo

College of Systems Engineering, National University of Defense Technology, Changsha 410073, China.

Laboratory for Big Data and Decision, National University of Defense Technology, Changsha 410073, China.

Innovation (Camb). 2025 May 12;6(6):100948. doi: 10.1016/j.xinn.2025.100948. eCollection 2025 Jun 2.

Intelligent decision-making (IDM) is a cornerstone of artificial intelligence (AI) designed to automate or augment decision processes. Modern IDM paradigms integrate advanced frameworks to enable intelligent agents to make effective and adaptive choices and decompose complex tasks into manageable steps, such as AI agents and high-level reinforcement learning. Recent advances in multimodal foundation-based approaches unify diverse input modalities-such as vision, language, and sensory data-into a cohesive decision-making process. Foundation models (FMs) have become pivotal in science and industry, transforming decision-making and research capabilities. Their large-scale, multimodal data-processing abilities foster adaptability and interdisciplinary breakthroughs across fields such as healthcare, life sciences, and education. This survey examines IDM's evolution, advanced paradigms with FMs and their transformative impact on decision-making across diverse scientific and industrial domains, highlighting the challenges and opportunities in building efficient, adaptive, and ethical decision systems.

智能决策（IDM）是人工智能（AI）的基石，旨在使决策过程自动化或增强决策过程。现代IDM范式整合了先进框架，以使智能主体能够做出有效且适应性强的选择，并将复杂任务分解为可管理的步骤，如AI主体和高级强化学习。基于多模态基础的方法的最新进展将视觉、语言和感官数据等多种输入模态统一到一个连贯的决策过程中。基础模型（FMs）在科学和工业领域已变得至关重要，改变了决策和研究能力。它们大规模的多模态数据处理能力促进了跨医疗保健、生命科学和教育等领域的适应性和跨学科突破。本综述考察了IDM的发展历程、基于基础模型的先进范式及其对不同科学和工业领域决策的变革性影响，突出了构建高效、适应性强且符合伦理的决策系统所面临的挑战和机遇。