The International Workshop on Industrial Physical AI

Call

---

Physical AI represents a paradigm shift that extends AI from purely digital spaces into the physical world, integrating fundamental physical laws, multimodal perception, and data-driven learning to build reliable, causal, and deployable intelligent systems for real-world industrial environments. By embedding physical constraints, dynamics, and causality into model design and inference, Physical AI achieves strong robustness against distribution shifts and high reliability for safety-critical industrial scenarios. Rooted in industrial intelligence, it has become a key enabler for smart manufacturing, autonomous industrial systems, and the next generation of industrial AI.

This workshop focuses on the intersection of Industrial Physical AI and Multimedia Technology, aiming to provide a focused forum for researchers and practitioners to discuss the latest advances, challenges, and opportunities in multimodal-driven physical AI for real-world industrial scenarios.

Industrial Physical AI integrates physical modeling, AI algorithms, and industrial domain knowledge to solve complex problems in industrial production, operation, and maintenance; while multimodal technology serves as the foundation for Industrial Physical AI, industrial physical systems generate massive multimodal data (e.g., visual images/videos of equipment, acoustic signals, vibration sensor data, temperature/humidity time series, production process text records), and effectively fusing these multimodal data is the key to realizing accurate perception, reliable decision-making, and intelligent control of industrial physical processes.

The scope of the workshop covers, but is not limited to, the following topics:

• Multimodal data acquisition, preprocessing, and representation for industrial physical systems
• Physical constraint-aware data fusion algorithms for industrial scenarios
• Multimodal-driven industrial fault diagnosis and prognosis
• Industrial physical simulation with multimodal feedback
• Human-machine interaction for industrial physical AI systems
• Edge-aware learning for industrial edge devices
• Benchmark datasets and evaluation metrics for industrial physical AI
• Real-world deployment cases of Industrial Physical AI