Guest Editor(s)
Dr. Xin Zhang
Email: mexzyl@ust.hk
Affiliation: Department of Mechanical and Aerospace Engineering, The Hong Kong University of Science and Technology, Hong Kong, China
Homepage:
Research Interests: complex equipment condition recognition, multimodal industrial large model
Dr. Xinming Li
Email: lxm18134462961@163.com
Affiliation: School of Mechanical-Electronic and Vehicle Engineering, Beijing University of Civil Engineering and Architecture, Beijing, China
Homepage:
Research Interests: graph neural network, intelligent diagnosis, multimodal industrial large model and structural health monitoring
Dr. Xi Zhang
Email: xi7zhang@polyu.edu.hk
Affiliation: Department of Industrial and Systems Engineering, The Hong Kong Polytechnic University, Hong Kong, China
Homepage:
Research Interests: digital twin, human-robot collaboration
Dr. Fan Zhang
Email: mafzhang@ust.hk
Affiliation: Department of Mathematis, The Hong Kong University of Science and Technology, Hong Kong, China
Homepage:
Research Interests: reinforcement learning,largelanguage model, AI4 science
Summary
Agentic foundation models are emerging as an important new paradigm in artificial intelligence. Compared with conventional systems that focus mainly on perception or prediction, these models integrate perception, reasoning, planning, tool use, and action execution within a unified framework. With the rapid progress of large language models, vision-language models, multimodal foundation models, and intelligent agents, this paradigm is creating new opportunities for industrial intelligence. Traditional industrial AI has largely concentrated on specific tasks such as fault detection, quality inspection, and condition monitoring. By contrast, agentic foundation models make it possible to support more advanced and closed-loop industrial intelligence in complex environments, including diagnosis, predictive maintenance, process optimization, scheduling, and autonomous decision support.
At the same time, the development and deployment of agentic foundation models in industrial settings still face important challenges. General-purpose foundation models often require substantial adaptation before they can meet the needs of industrial tasks. High-quality industrial datasets remain limited, and the integration of foundation models with industrial software, operational workflows, and physical systems is still far from straightforward. In addition, industrial applications place strict requirements on safety, robustness, interpretability, trustworthiness, and real-time responsiveness. Many scenarios also involve heterogeneous multimodal data and require reliable coordination across devices, production lines, and factories. These issues call for further research on model design, multimodal learning, knowledge integration, agent coordination, and deployment frameworks oriented toward real industrial environments.
This special issue aims to collect high-quality contributions on the theory, methods, and applications of agentic foundation models for industrial intelligence. It will provide a forum for researchers and practitioners to report recent advances in industrial AI systems that are capable of perception, reasoning, planning, and autonomous decision-making in real-world scenarios.
Topics of interest include, but are not limited to:
· Agentic foundation models for industrial perception, reasoning, planning, and decision-making
· Large language models, vision-language models, and multimodal foundation models for industrial intelligence
· Industrial agents for diagnostics, predictive maintenance, scheduling, quality inspection, and process control
· Multi-agent collaboration and coordination in industrial systems
· Tool-augmented industrial agents based on retrieval, knowledge graphs, function calling, and industrial software interfaces
· Multimodal data fusion for industrial monitoring, anomaly detection, and root-cause analysis
· Closed-loop industrial intelligence from sensing to decision and action
· Foundation models and agents for digital twins, cyber-physical systems, and industrial Internet of Things applications
· Pre-training, fine-tuning, adaptation, and alignment strategies for industrial foundation models
· Safety, robustness, interpretability, trustworthiness, and human-in-the-loop collaboration for industrial agents
· Edge deployment, lightweight industrial agents, and cloud-edge-end collaboration
· Benchmark datasets, evaluation protocols, and industrial case studies for agentic AI
Keywords
agentic foundation models, industrial intelligence, autonomous decision-making, multimodal learning, industrial AI agents, predictive maintenance, digital twins
Login
Register
Submit a Paper
Propose a Special lssue