Visual world model is a cutting-edge direction in the current field of artificial intelligence. Taking visual signals as the core driving force, it enables intelligent systems to learn spatio-temporal rules, dynamic evolution and causal logic of the physical world from images and videos, so as to realize the deduction of future scenes and autonomous decision-making. It is a critical path towards general visual intelligence. Against this backdrop, visual world models have drawn increasingly widespread attention. To report the latest research achievements in this field and promote the coordinated development of visual world models in terms of theory, technology, security and implementation, the Visual Intelligence journal officially launches the call for papers for the thematic issue focused on "Visual World Model". Scholars in relevant fields are warmly welcome to submit their manuscripts.

1. Scope

The topics include but are not limited to the following areas:

1. Basic Theories and Representation Systems

• Unified representation framework for vision-driven world models
• Spatio-temporal coherence modeling methods across time sequences and perspectives
• Visual world modeling theories integrated with physical priors
• Causal reasoning and rule mining mechanisms for visual data

2. Core Architectures and Key Algorithms

• Long-sequence high-precision visual future state prediction models
• Generalization optimization methods under few-shot and weak-supervision conditions
• Deployment technologies for lightweight visual world models on edge devices
• World model architecture innovation integrated with vision-language large models

3. Security Challenge and Adversarial Attack & Defense

• Adversarial vulnerability mechanism and vulnerability analysis of visual world models
• Targeted adversarial attack technologies in digital and physical scenarios
• Efficient and low-overhead defense schemes for visual world models
• Design methods for provably robust safety alignment models

4. Cutting-edge Technical Directions of Visual World Models

• 3D/4D dynamic visual world modeling
• Visual world interaction models in embodied intelligence scenarios
• Fusion of controllable visual content generation and world models
• Visual world modeling in multi-agent collaborative scenarios

5. Evaluation System and Benchmark Construction for Visual World Models

• Specialized evaluation indicator system for visual world models
• Construction of public benchmark datasets for visual world models targeting different scenarios
• Interpretability evaluation methods for visual world models
• Special evaluation benchmarks for adversarial attack and defense of visual world models

2. Important Dates

Submission Deadline: 31 December 2026

First Round Review Comments Released: 20 February 2027

Author Revision Submission Deadline: 30 March 2027

Final Review Comments Released: 5 May 2027

Publication Date: 30 June 2027

3. Submission Guidelines

Submitted manuscripts can be new relevant research achievements, or extended versions of conference papers (the extended conference papers are required to add no less than 50% new research content). All submitted manuscripts will receive review comments from at least 3 peer experts in accordance with the requirements of Visual Intelligence. The journal will decide whether to accept the manuscript based on its quality, innovation of the research topic, and compliance with the theme of this thematic issue. All manuscripts shall be written in English with no less than 10 pages.

How to Submit:

Please visit Visual Intelligence at https://link.springer.com/journal/44267

Submit your manuscript and click the Submit your manuscript button.

Notes:

Please select Thematic Series when submitting your manuscript;

Select the thematic series ‌Visual World Model‌ for submission.

4. Guest Editors

Fernando Basura Ph.D

Agency for Science, Technology and Research (A*STAR), Singapore, Singapore.

Dr. Basura Fernando is a Principal Scientist and NRF Fellow at A*STAR, Singapore, and a Principal Investigator at the Centre for Frontier AI Research (CFAR), Institute of Artificial Intelligence and Computing (IAIC). He also serves as an Adjunct Assistant Professor at the College of Computing and Data Science, Nanyang Technological University (NTU). Previously, he was an Honorary Lecturer and Research Fellow at the Australian National University (ANU) and the Australian Centre for Robotic Vision (ACRV). He received his Ph.D from KU Leuven, Belgium.

Qing Guo Ph.D

College of Computer Science, Nankai University, Tianjin, China.

Dr. Qing Guo is a Full Professor at Nankai University, China. He serves as Area Chair for NeurIPS, ICML, ICLR, ICCV, CVPR, and AAAI, and as Associate Editor of IEEE Transactions on Image Processing, TMLR, and Visual Intelligence, among others. His research focuses on computer vision, AI safety, adversarial attacks, and robustness, with over 100 papers published in top-tier venues including ICML, NeurIPS, ICLR, CVPR, ICCV, SP, CCS, TPAMI, and IJCV.

Yi Jin Ph.D

Professor, School of Computer Science, Beijing Jiaotong University, Beijing, China.

Dr. Jin Yi, Professor at the School of Computer Science, Beijing Jiaotong University. She is a Council Member of China Computer Federation (CCF). She is Vice Director of the Key Laboratory of Transportation Big Data and Artificial Intelligence of the Ministry of Education of China, and member of the National Key R&D Program and the National Science and Technology Major Special Project Expert Group. She received the Excellent Achievement Award from the MOE of China, the Best Paper Nomination Award from IEEE CS, and the Railway Science and Technology Progress Award. Her research focuses on multimodal data processing and fusion perception.

Jun Liu Ph.D

School of Computing and Communications, Lancaster University, Lancaster, United Kingdom.

Dr. Jun Liu is Professor at the School of Computing and Communications, Lancaster University. He is a Fellow of International Association for Pattern Recognition (IAPR) and a Fellow of British Computer Society (BCS). He serves as Associate Editor-in-Chief for Pattern Recognition and Senior Area Editor for IEEE Transactions on Image Processing. He received the Best Associate Editor Award of IEEE TBIOM in 2026. He is General Chair of BMVC 2026, and Program Chair of BMVC 2025 and VCIP 2026. His research focuses on AI, computer vision, machine learning, and digital health applications.

Qing Zhang Ph.D

School of Computer Science and Engineering, Sun Yat-sen University, Guangzhou, China.

Dr. Qing Zhang is an Associate Professor with the School of Computer Science and Engineering, Sun Yat-sen University. His research interests include world model, physical AI, and embodied AI. He has now published more than 80 papers in journals (including TPAMI, IJCV, and TOG) and conferences (including ICCV, CVPR, SIGGRAPH, and NeurIPS).