Workshop 1Submission Link for Workshop 1
Summary:
Real-world data, such as images, networks, and text, inherently possess multi-granularity structural characteristics. Inspired by human multi-granularity cognitive thinking, multi-granularity cognitive computing has emerged as an advanced computational paradigm that enables modeling and solving complex problems at different granularities and scales. By integrating multi-granularity analysis and cognitive computing, it is possible to overcome the limitations of traditional data mining and achieve more intelligent and efficient mining.
Additionally, current machine learning models often present as "black boxes," lacking interpretability and understandability, which limits their controllability and widespread deployment in real-world applications. Although multi-granularity cognitive computing models have demonstrated powerful modeling and problem-solving capabilities, their complexity may pose challenges in terms of interpretability. Moreover, the inherent fuzziness and uncertainty in multi-granularity cognitive computing models can further complicate their interpretability and reliability. Therefore, investigating ways to enhance the interpretability of models, address their fuzziness and uncertainty, and make their decision-making processes more transparent and understandable is crucial for developing trustworthy and reliable artificial intelligence.
This workshop aims to bring together researchers, practitioners, and industry experts to discuss the latest developments, challenges, and future directions of multi-granularity cognitive computing in data mining. We invite submissions of original research papers, and position papers that address the following topics, but are not limited to:
Additionally, current machine learning models often present as "black boxes," lacking interpretability and understandability, which limits their controllability and widespread deployment in real-world applications. Although multi-granularity cognitive computing models have demonstrated powerful modeling and problem-solving capabilities, their complexity may pose challenges in terms of interpretability. Moreover, the inherent fuzziness and uncertainty in multi-granularity cognitive computing models can further complicate their interpretability and reliability. Therefore, investigating ways to enhance the interpretability of models, address their fuzziness and uncertainty, and make their decision-making processes more transparent and understandable is crucial for developing trustworthy and reliable artificial intelligence.
This workshop aims to bring together researchers, practitioners, and industry experts to discuss the latest developments, challenges, and future directions of multi-granularity cognitive computing in data mining. We invite submissions of original research papers, and position papers that address the following topics, but are not limited to:
- Algorithms and models for multi-granularity cognitive computing in data mining.
- Interpretability, understandability, and robustness of multi-granularity cognitive computing models
- Multi-granularity representation learning and feature extraction methods.
- Integration of domain knowledge and cognitive mechanisms in data mining to handle fuzziness and uncertainty
- Applications of multi-granularity cognitive computing in graph data mining, natural language processing, computer vision, healthcare, social network analysis, and other domains.
- Evaluation metrics and benchmarks for assessing the interpretability, robustness, and reliability of models under fuzziness and uncertainty.
Keywords: Multi-granularity cognitive computing, Data mining, Model interpretability, Fuzziness and uncertainty, Representation learning, Domain knowledge integration, Evaluation metrics and benchmarks, , Natural language processing, Graph neural networks, Computer vision.
Research interests: Data Mining, Knowledge Graph, Graph Neural Networks, Model Explainability.
2019-present: Associate Professor, School of Computer Science and Technology, Chongqing University of Posts and Telecommunications, China
2021-2023: Post-Doc at Hong Kong Baptist University supported by the HK Scholar Programme.
2016-2019: Assistant Professor, School of Computer Science and Technology, Chongqing University of Posts and Telecommunications, China
2012-2016: Beijing Institute of Technology, PhD in Engineering
2019-present: Associate Professor, School of Computer Science and Technology, Chongqing University of Posts and Telecommunications, China
2021-2023: Post-Doc at Hong Kong Baptist University supported by the HK Scholar Programme.
2016-2019: Assistant Professor, School of Computer Science and Technology, Chongqing University of Posts and Telecommunications, China
2012-2016: Beijing Institute of Technology, PhD in Engineering
Research interests: Uncertainty Data Mining, Knowledge Graph, Graph Neural Networks, Model Explainability.
2009-present: Professor, School of Computer Science and Technology, Chongqing University of Posts and Telecommunications, China
2003-2009: Associate Professor, School of Computer Science and Technology, Chongqing University of Posts and Telecommunications, China
2005-2008: Chongqing University, PhD in Computer Science
2009-present: Professor, School of Computer Science and Technology, Chongqing University of Posts and Telecommunications, China
2003-2009: Associate Professor, School of Computer Science and Technology, Chongqing University of Posts and Telecommunications, China
2005-2008: Chongqing University, PhD in Computer Science
Workshop 2Submission Link for Workshop 2
Summary:
Artificial Intelligence (AI) is increasingly being deployed in real-world systems and applications that impact critical aspects of society, including healthcare, transportation, cybersecurity, industrial automation, and smart infrastructure. As AI systems become more autonomous, interconnected, and influential, ensuring their trustworthiness has emerged as a fundamental challenge. Key concerns include robustness, security, privacy, explainability, transparency, reliability, and responsible deployment.
This workshop aims to bring together researchers and practitioners from academia and industry to explore recent advances in trustworthy AI from the perspectives of methods, systems, and applications. We welcome contributions on foundational techniques for building trustworthy AI, system-level approaches for deploying AI in complex and dynamic environments, and application-driven studies that demonstrate the practical impact of trustworthy AI technologies.
Topics of interest include, but are not limited to:
The workshop seeks to foster interdisciplinary collaboration and identify future research directions toward intelligent IoT systems that are secure, reliable, transparent, and capable of operating safely in the physical world.
This workshop aims to bring together researchers and practitioners from academia and industry to explore recent advances in trustworthy AI from the perspectives of methods, systems, and applications. We welcome contributions on foundational techniques for building trustworthy AI, system-level approaches for deploying AI in complex and dynamic environments, and application-driven studies that demonstrate the practical impact of trustworthy AI technologies.
Topics of interest include, but are not limited to:
- Trustworthy and Explainable AI
- AI Security and Adversarial Robustness
- Privacy-Preserving Machine Learning
- Federated and Collaborative Learning
- LLM and Agentic AI
- AI-Generated Content Detection
- AI for Network Traffic Analysis and Representation Learning
- Trustworthy AI applications in healthcare, intelligent transportation, and other real-world domains.
The workshop seeks to foster interdisciplinary collaboration and identify future research directions toward intelligent IoT systems that are secure, reliable, transparent, and capable of operating safely in the physical world.
Keywords: Trustworthy AI, AI Security and Safety, Internet of Things (IoT), Agentic AI
Edward Szczerbicki received the D.Sc. degree in information science from the Szczecin University of Technology, Szczecin, Poland, in 1993. He had very extensive experience in the area of intelligent systems development over an uninterrupted 40-year period, 25 years of which he spent in top systems research centers in the United States, U.K., Germany, and Australia. In this area, he contributed to the understanding of information and knowledge management in systems operating in environments characterized by informational uncertainties. He has published close to 350 refereed papers with over 2000 citations over the last 20 years. His academic experience includes ongoing positions with Gdansk University of Technology, Gdansk, Poland; Strathclyde University, Glasgow, Scotland; the University of Iowa, Iowa City, IA, USA; the University of California at Berkeley, Berkeley, CA, USA; and the University of Newcastle, Callaghan, NSW, Australia. He has given numerous invited presentations and addresses at universities in Europe, USA, and at international conferences. He received the Title of Professor of information science for his international published contributions in 2006. Prof. Szczerbicki serves as a Board Member of Knowledge Engineering Systems, and a Member of Berkeley Initiative in Soft Computing Special Interest Group on Intelligent Manufacturing. He is a Member of the Editorial Board/Associated Editor for eight international journals. He chaired/co-chaired and acted as a committee member for several international conferences.
Prof. Haoxi Zhang is an Associate Professor at Chengdu University of Information Technology, China. He holds a Ph.D. in Knowledge Engineering from the University of Newcastle, Australia (2013), and a Master閳ユ獨 in Software Engineering from the University of Electronic Science and Technology of China. Prof. Zhang's research intricately melds artificial intelligence with healthcare, focusing on the development of advanced computational models and machine learning algorithms for biomedical data analysis, particularly medical imaging and AIoT for healthcare, to enhance medical decision-making. His work emphasizes developing multimodal learning algorithms to integrate multi-scale biomedical data for comprehensive disease management, constructing real-world learning systems for creating robust, trustworthy representations from imperfect medical data, and innovating causality-driven learning algorithms to boost interpretability and safety in healthcare applications. Prof. Zhang has published over 40 refereed papers in leading journals and conferences.
Workshop 3Submission Link for Workshop 3
Summary:
Artificial Intelligence is experiencing a transformative era, predominantly driven by rapid breakthroughs in Natural Language Processing (NLP) and Computer Vision (CV). As these two foundational pillars of AI continue to evolve and increasingly intersect—particularly in the rapidly advancing realm of multimodal intelligence—the need for robust methodologies, comprehensive benchmarks, and accurate evaluation metrics has never been more critical.
This workshop aims to provide a premier interdisciplinary forum for researchers, practitioners, and academicians to present and discuss the most recent frontier advances across both NLP and CV. We invite original contributions that explore novel algorithms, theoretical insights, practical applications, and newly constructed datasets or evaluation protocols. By fostering a collaborative dialogue between the language and vision communities, this workshop seeks to bridge existing disciplinary gaps, inspire innovative problem-solving, and catalyze efforts that will shape the future of generalizable and robust AI systems.
This workshop aims to provide a premier interdisciplinary forum for researchers, practitioners, and academicians to present and discuss the most recent frontier advances across both NLP and CV. We invite original contributions that explore novel algorithms, theoretical insights, practical applications, and newly constructed datasets or evaluation protocols. By fostering a collaborative dialogue between the language and vision communities, this workshop seeks to bridge existing disciplinary gaps, inspire innovative problem-solving, and catalyze efforts that will shape the future of generalizable and robust AI systems.
Keywords:
Natural Language Processing (NLP)
Computer Vision (CV)
Multimodal Intelligence
Bowen Xing is an associate professor at school of computer and communication engineering, University of Science and Technology Beijing. He received his B.E. degree and Master degree in computer science from Beijing Institute of Technology in 2017 and 2020, respectively. He obtained his PhD degree in artificial intelligence in 2024, from Australian Artificial Intelligence Institute (AAII), University of Technology Sydney (UTS), under the supervision of Professor Ivor W. Tsang. His research focuses on artificial intelligence, natural language processing, large langue models, and knowledge-enhanced inference. He has published papers on top-tier conferences and journals (e.g., IEEE TPAMI, JAIR, ACL, EMNLP, IJCAI, ECML). He serves as Area Chair, PC member and reviewer of ACL rolling review, EMNLP, AAAI, IJCAI, AISTATS, IEEE TNNLS, IEEE TASLP, etc.
Workshop 4Submission Link for Workshop 4
Summary:
In many scenarios such as target recognition and tracking, defect detection, image and text generation, intelligence analysis, and intelligent Q&A, it is difficult to ensure the accuracy of recognition, detection, or answering using a single modal data modeling due to complex and variable environments, strong background noise, and insufficient information. To improve the modeling effect, it is necessary to fully utilize the complementary advantages of multi-modal data, integrate data from various modalities such as RGB images, infrared images, ultrasound, text, and electromagnetic signals, and construct a unified feature representation model. This workshop will focus on several academic issues in multi-modal fusion in artificial intelligence, including but not limited to: object recognition and tracking, defect detection, image and text generation, intelligence analysis, intelligent Q&A, RAG, Agent, and other aspects.
Keywords: Multi-modal fusion, Artificial intelligence, Feature representation, Large models, RAG
As the project leader or main participant, Professor Chen participated in many projects such as 863, NSFC. In recent years, He has published more than 130 academic papers in many important academic journals or international conferences such as Physics Report, Knowledge Based Systems, Scientific Data and Information Sciences. He Received the second prize of the Natural Science Award of the CCF in 2014, the third prize of the Sichuan Provincial Science and Technology Progress Award in 2022, and the Second Prize of the Wu Wenjun Artificial Intelligence Science and Technology Award in 2025.
Hongliang SUN mainly focuses on graph computing and complex networks using graph neural networks to tackle real-world problems. He has published more than 30 papers on PNAS nexus, WWW, KBS and etc. He also serves as the senior member of CCF. He has also been granted multiple projects, including the National Natural Science Foundation of China and the JiangSu Provincial Natural Science Foundation.
Yan-Li Lee mainly engaged in research at graph mining and natural language processing, focusing on basic theories and algorithms in the field of graph mining and natural language processing, as well as socioeconomic problems that can be solved using graph mining research paradigms and natural language processing technology. She serves as a member of the Specialized Committee on Social Media Processing of the Chinese Information Society of China, a member of the Women Working Committee of the China Computer Federation，the deputy secetary-general of the Specialized Committee on Natural Language Processing of the Sichuan Computer Society, and the associate deputy director of the National First-Class Computer Science and Technology Program at Xihua University. She has published more than 40 papers in international journals such as Pattern Recognition, Knowledge Based Systems, PNAS Nexus, Information Fusion, Applied Mathematics and Computation, and Physica A. She has also been granted multiple projects, including the National Natural Science Foundation of China and the Sichuan Provincial Natural Science Foundation.
Workshop 5Submission Link for Workshop 5
Summary:
Recent advances in computer vision and multimodal artificial intelligence have enabled visual perception systems to move beyond conventional category-level object detection toward the detection and localization of specific targets described by natural language, reference images, semantic attributes, or contextual relationships. Unlike traditional object detection, which identifies all instances belonging to predefined categories, specific object detection aims to accurately locate the particular object that best matches a given user intention or multimodal query. This workshop focuses on the theories, methods, datasets, and applications of specific object detection and localization in complex visual scenes. It will provide a platform for researchers and practitioners to discuss recent progress in vision-language grounding, referring expression comprehension, open-vocabulary detection, reference-based object localization, multimodal feature alignment, and fine-grained visual understanding. The workshop welcomes original research on multimodal representation learning, foundation models, efficient adaptation, prompt-based detection, few-shot and zero-shot localization, spatial relationship reasoning, dataset construction, evaluation protocols, and trustworthy visual perception.
Keywords: Specific Object Detection; Object Localization; Vision-Language Grounding; Referring Expression Comprehension;
Liujie Hua received the PhD degree from Central South University. Currently, he is a lecturer in the Department of Criminal Science and Technology at Hunan Police Academy. His research interests mainly include multimodal data processing, UAV low-altitude inspection, and related forensic intelligent technologies. He has published more than 30 papers in prestigious conferences and journals, including AAAI, CVPR, and EMNLP.
Miaojiang Chen received the Ph.D. degree in computer science from Central South University in 2023. He is currently an Associate Professor of School of Computer and Electronic Information, Guangxi University, China. He has published several journal and conference papers in the JSAC, TON, TMC, TSC, AAAI, TITS, TNSE, TETCI, TCE, TOMM, etc., and he also serves reviewer of the top-tier conferences and journals, including ICML, IEEE Transactions on Parallel and Distributed Systems, IEEE Trans. on Information Forensics and Security, IEEE trans. on industrial informatics, IEEE Trans. on Intelligent Transportation Systems, IEEE Internet of things journal. He won the IEEE HITC 2025 Award for Excellence in Hyper-Intelligence (Early Career Researchers), and Young Talents of the Guangxi High-Level Personnel Special Support Program. His major research interests include deep reinforcement learning, Internet of Things, edge computing, transfer learning, optimization.
Workshop 6Submission Link for Workshop 6
Summary:
Image Computing refers to the technology that utilizes computer to process, analyze, and understand image data, aiming to extract meaningful information, achieve specific functions, or solve practical problems.
With the development of deep learning and large models, image computing technologies have made significant progress. This session highlights advances and applications in image computing research.
Keywords:
The topics to be covered include, but are not limited to:
- Image classification
- Image retrieval
- Image registration
- Image segmentation
- Image captioning
- Image watermarking
- Image compression
- Image generation
- Image reconstruction
- Image enhancement
- Image denoising
- Image super-resolution
- Image restoration
- Image object detection
- Image anomaly detection
- Image quality assessment
- Image depth estimation
- Image feature extraction
- Image feature learning
- Image privacy protection
- Image dataset generation
- Image computing systems
- Image feature representation: Image-level feature representation, dataset-level feature representation
- Image-to-text visual question answering
- Image dataset quality assessment
- Multimodal image fusion
- Multi-view image reconstruction
- Lightweight image computing models
- Medical image processing
- Remote sensing image processing
- Security and ethical issues in image computing
- Large model-based image computing methods
Huanjie Tao received the M.S. degree in Mathematics and Information Science from Capital Normal University (CNU), Beijing, China in 2016, and received the Ph.D. degree in pattern recognition and intelligent system from Southeast University (SEU), Nanjing, China in 2020. He was a visiting PhD student at the School of Computer Science and Engineering at Nanyang Technological University (NTU) from September 2018 to September 2019. He is currently employed as an associate professor with the School of Computer Science, Northwestern Polytechnical University (NPU), Xi’an, China. He also works in the Engineering Research Center of Embedded System Integration, Ministry of Education, and the National Engineering Laboratory for Integrated Aero-Space-Ground-Ocean Big Data Application Technology. His research includes large model and deep learning.
Workshop 7Submission Link for Workshop 7
Summary:
Traditional multimodal fusion methods typically rely on predefined categories and closed-set assumptions, excelling in controlled benchmarks but faltering in real-world deployments. In open-world environments, systems must contend with continuously emerging unknown categories, dynamic modality reliability, and unpredictable sensor failures - scenarios where conventional fusion strategies often break down. Moreover, existing black-box integration mechanisms offer little interpretability into how modalities interact or why certain fusion decisions are made.
This workshop addresses the emerging paradigm of openness-aware multimodal fusion - a framework that explicitly accounts for both unknown categories and the inherent openness of real-world data distributions. Recent advances have pioneered pseudo-unknown sample generation and perception-augmented open-set training to simultaneously enhance generalization and interpretability. Concurrently, methods demonstrate that explicit modeling of multimodal interactions can yield not only performance gains but also local and global interpretability. Open-set cross-modal generalization further challenges models to transfer knowledge across modalities while generalizing to unseen classes.
We invite contributions spanning openness-aware fusion architectures, interpretable multimodal reasoning, open-set recognition, robustness to modality missing/noise, and theoretical foundations for open-world multimodal learning. By bridging the gap between closed-set precision and open-world adaptability, this workshop aims to chart a path toward multimodal AI systems that are both robust and trustworthy in the wild.
This workshop addresses the emerging paradigm of openness-aware multimodal fusion - a framework that explicitly accounts for both unknown categories and the inherent openness of real-world data distributions. Recent advances have pioneered pseudo-unknown sample generation and perception-augmented open-set training to simultaneously enhance generalization and interpretability. Concurrently, methods demonstrate that explicit modeling of multimodal interactions can yield not only performance gains but also local and global interpretability. Open-set cross-modal generalization further challenges models to transfer knowledge across modalities while generalizing to unseen classes.
We invite contributions spanning openness-aware fusion architectures, interpretable multimodal reasoning, open-set recognition, robustness to modality missing/noise, and theoretical foundations for open-world multimodal learning. By bridging the gap between closed-set precision and open-world adaptability, this workshop aims to chart a path toward multimodal AI systems that are both robust and trustworthy in the wild.
Keywords: Multimodal Fusion, Multi-view Learning, Open-set Recognition, Out-of-distribution Detection.
Shiping Wang is currently working as a Full Professor with the College of Computer and Data Science, and the director of the Fujian Provincial Key Laboratory of Intelligent Metro, Fuzhou University, Fuzhou, China. He has authored or co-authored over 200 papers with more than 8000 citations in top conferences and journals, including T-PAMI, TIP, TMM, TSMC, TSP, TITS, TCSVT, TKDD, CVPR, AAAI, ACM MM, PR, NN, INS, KBS, etc. He has severed as the Regional Chair of the 10th Pacific-Rim Symposium on Image and Video Technology and IEEE International Conference on Digital Twins and Parallel Intelligence (DTPI), General Chair of the 2023 IEEE 6th International Conference on Pattern Recognition and Artificial Intelligence (PRAI 2023) and the Session Chair of the 2023 IJCAI Young Elite Symposium (IJCAI YES 2023).
Sheng Lian is currently working as a Lecturer and Master’s Supervisor with the College of Computer and Data Science, Fuzhou University, Fuzhou, China. He is a recipient of the Fujian Provincial High-Level Talent. His research interests include intelligent medical image analysis, computer vision, artificial intelligence, and machine learning. He has published in top venues (e.g., IJCV, MedIA, Neural Networks, Brain, IEEE TMM/JBHI/TCBB, Remote Sensing, MICCAI, ICME, BIBM) with 2,000+ Google Scholar citations, and has led or participated in multiple funded projects, including NSFC Youth and General Programs, Ministry of Education projects, and Fujian Provincial Natural Science Foundation.
Workshop 8Submission Link for Workshop 8
Summary:
Rapid advances in artificial intelligence, high-throughput sequencing, medical imaging, electronic health records, and real-world data are transforming biomedical research and clinical decision-making. This workshop focuses on computational and AI-driven methods for integrating, mining, and interpreting large-scale biomedical data. Topics include multimodal and multi-omics data integration, single-cell and spatial omics, computational pathology and medical imaging, precision medicine, clinical data mining, drug-target interaction prediction and AI-driven drug discovery, biomedical foundation models, graph learning and generative AI, and trustworthy and interpretable medical AI. The workshop aims to bring together researchers from computer science, bioinformatics, biomedical engineering, pharmacy, and clinical medicine to exchange methodological advances, benchmark practices, and translational applications, with particular emphasis on robust, reproducible, and clinically meaningful biomedical big-data analytics.
Keywords: Biomedical Big Data; Artificial Intelligence; Multi-Omics; Single-Cell and Spatial Omics; Computational Pathology; Precision Medicine; AI-Driven Drug Discovery
Dr. Siwen Xu is a Lecturer at the School of Medical Information Engineering, Guangdong Pharmaceutical University, China. He received his Ph.D. in Bioinformatics and has conducted postdoctoral research in biomedical informatics and precision medicine. His research focuses on artificial intelligence and biomedical big-data analytics, including multi-omics integration, single-cell and spatial omics, computational pathology, regulatory genomics, and AI-driven drug discovery. He has developed computational methods and software for single-cell demultiplexing, regulatory element analysis, epigenomic data analysis, and cancer prognosis prediction. His work has been published in journals including Briefings in Bioinformatics, IEEE Journal of Biomedical and Health Informatics, BMC Genomics, and Frontiers in Bioinformatics.
Dr. Zixiao Lu is an Associate Researcher Fellow at the School of Medical Information Engineering, Guangzhou University of Chinese Medicine, China. She received her Ph.D. degree in Biomedical Engineering and has extensive research experience in artificial intelligence, bioinformatics, and biomedical big-data analytics. Her research focuses on AI-driven precision medicine, including medical image analysis, multi-omics integration, computational pathology, spatial omics, and AI-driven drug discovery. She has developed computational methods and software for medical image component segmentation, multi-omics integration, cancer prognosis prediction, gene co-expression network analysis, and single-cell genomics. Her work has been published in journals including Nature Communications, Medical Image Analysis, Genomics, Proteomics & Bioinformatics, Briefings in Bioinformatics, IEEE Journal of Biomedical and Health Informatics, and BMC Genomics. She has also been granted multiple Chinese invention patents for AI- and bioinformatics-based biomedical data analysis methods.
Dr. Jiahong Wang is an Associate Researcher Fellow at the School of Basic
Medical Sciences, Southern Medical University. He received his Ph.D. degree
in Oncology and has extensive research experience in bioinformatics,
and biomedical big-data analytics. His research focuses on the development
of bioinformatics analysis tools, literature mining for gene function and
gene regulatory networks, and multi-omics integration. His work has been
published in journals including Nature Communications, Bioinformatics,
Autophagy, Scientific Reports, Molecular Cancer Research, and Head & Neck.
Weizhi Nie is a Professor and Master's/Ph.D. supervisor in the School of Electrical and Information Engineering at Tianjin University. His research focuses on artificial intelligence, intelligent healthcare, AI agents, large language models, multimedia information analysis, and cross-media information retrieval. His work has appeared in leading venues including IEEE TPAMI, IEEE TMI, IEEE TMM, IEEE TNNLS, IEEE TKDE, IEEE TBME, and MICCAI. He has served as principal investigator for three projects funded by the National Natural Science Foundation of China and one Tianjin Natural Science Foundation project, and as a subproject leader in the National Key R&D Program of China. His honors include the 2025 First Prize for Science and Technology Progress from the China General Chamber of Commerce, the 2025 Special Prize for Teaching Achievement of Guangdong Province, and the 2019 and 2020 Second Prizes for Science and Technology Progress of Tianjin.
Workshop 9Submission Link for Workshop 9
Summary:
While the convergence of Big Data and AI drives transformative
progress, it also introduces critical security and privacy
challenges—from data-scale vulnerabilities to model complexity.
Beyond these, the growing deployment of autonomous intelligent agents
(e.g., LLM-based agents, multi-agent systems) adds novel risks: goal
manipulation, tool abuse, reasoning leakage, and impersonation across
agent interactions. Collaborative multi-agent settings further amplify threats like
coordination poisoning
and aggregate privacy erosion.
This workshop invites contributions on fundamental principles and practical techniques for building secure, transparent, and privacy-preserving intelligent systems, with particular focus on agent-centric threat modeling, verifiable safety guarantees, privacy-aware communication protocols, and accountability frameworks for autonomous decision-making.
This workshop invites contributions on fundamental principles and practical techniques for building secure, transparent, and privacy-preserving intelligent systems, with particular focus on agent-centric threat modeling, verifiable safety guarantees, privacy-aware communication protocols, and accountability frameworks for autonomous decision-making.
Keywords: Privacy, Security, Artificial Intelligence, Distributed System
Tian Zhou is an Assistant Professor at Xi’an Jiaotong
University. He holds dual Ph.D. degrees from Xi'an Jiaotong
University and University of Massachusetts Amherst.
His primary research interests include distributed computing,
artificial intelligence, and privacy protection.
He has published over ten academic papers in top-tier
journals and conferences such as WWW, TPDS, ICDCS, and
BigData, including a Best Paper Runner-up Award at IEEE
IC2E 2021. He also serves as a reviewer and committee
member for multiple international journals and conferences,
including TPDS, IEEE Network, TrustCom and so on.
Linkang Du is an Assistant Professor at Xi’an Jiaotong University,
specializing in trustworthy artificial intelligence. His research focuses on data security,
privacy protection, and dataset copyright auditing in AI systems. He received his Ph.D. from
Zhejiang University, with additional research experience at CISPA Helmholtz Center for Information
Security in Germany and the Singapore University of Technology and Design. His work has been published
at top venues such as IEEE S&P, NDSS, ACM CCS, USENIX Security, and ACM WWW. Dr. Du is committed to
safeguarding data rights in the era of widespread AI adoption.
Workshop 10Submission Link for Workshop 10
Summary:
The proliferation of latency-sensitive and compute-intensive applications, ranging from autonomous driving and augmented reality to industrial IoT, has catalyzed the rapid deployment of Mobile Edge Computing (MEC) and multi-access edge networks. However, the stringent Quality-of-Service (QoS) requirements, resource heterogeneity, and dynamic network conditions pose significant challenges to efficient edge network operation. This workshop aims to bring together researchers and practitioners from academia and industry to discuss advances in optimization theory, algorithm design, and system implementation for mobile edge networks. We welcome original contributions that address theoretical foundations, novel architectures, performance analysis, and real-world testbed evaluations.
Topics of interest include, but are not limited to:
Topics of interest include, but are not limited to:
- joint computation offloading and resource allocation
- proactive and reactive edge caching strategies
- network slicing and virtualization for edge ecosystems
- energy-efficient edge orchestration
- AI/ML-driven network optimization.
Keywords: Mobile Edge Computing; Computation Offloading; Edge Caching; Network Slicing; Resource Allocation; Distributed Intelligence; QoS-Aware Scheduling
Chuanfen Feng received the B.E. degree from Ludong University, Yantai, China, in 2002, and the Ph.D. degrees from Beijing University of Posts and Telecommunications, Beijing, China, in 2007. He is currently a professorate senior engineer with the School of Information Science and Engineering, Shandong Normal University. His current research interests include mobile network planning, mobile edge computing and internet of things.
Yantong Wang received the BE degree in Internet of Things from Shandong University, China, in 2014, the MEng degree in Electronics and Communication from Tsinghua University, China, in 2017, and the PhD degree in telecommunications from King's College London, UK, in 2022. He is currently a lecturer at the School of Information Science and Engineering, Shandong Normal University, China. His research interests include proactive caching, edge network optimization, mathematical programming, and next-generation network technologies.
Workshop 11Submission Link for Workshop 11
Summary:
The rapid proliferation of large language models (LLMs) and foundation models is reshaping the landscape of cyberspace—offering unprecedented capabilities in data understanding, reasoning, and automation, while simultaneously introducing new attack surfaces, vulnerabilities, and governance challenges. Conversely, cyberspace security provides both a critical application domain and a rigorous testing ground for robust, trustworthy AI. This workshop focuses on the bidirectional empowerment between large models and cybersecurity: how large models can be leveraged to defend, detect, and respond to cyber threats, and how security principles (e.g., privacy preservation, adversarial robustness, and formal verification) can in turn guide the design, training, and deployment of safer, more resilient large models.
We invite original research, position papers, and case studies that bridge these two fast‑evolving fields.
Topics of Interest (including but not limited to)
We invite original research, position papers, and case studies that bridge these two fast‑evolving fields.
Topics of Interest (including but not limited to)
- LLM-driven threat detection and anomaly recognition
- Intelligent attack path reasoning and predictive defense
- LLM-enhanced automated penetration testing and red/blue team exercises
- LLM-enhanced cyber vulnerability discovery, prioritization, automated verification, and intelligence analysis
- LLM-assisted cyber vulnerability patching and remediation strategies
- Robustness and adversarial defense in LLM
- Data integrity and privacy-preserving learning in LLM
- Verification, validation, and testing of LLM security
- Risk identification, assessment, and mitigation in LLM systems
- Case studies and real-world deployment experiences in LLM security and risk management
- Security of AI Agent and Agentic systems
- Cyber deception and active defense
- Software and AI supply chain security
- AI-enhanced program analysis and software testing
Keywords: LLM-enhanced cyber security, LLM inherent security, LLM adversarial and defense
Hui Lu is a national-level distinguished talent, and currently serves as Dean of the Institute of Cyberspace Advanced Technology at Guangzhou University, and Director of the Guangdong Key Laboratory of Cyber Security for Critical Infrastructure. He holds the positions of Secretary-General of the China Cyberspace Emerging Technology Security Innovation Forum and Head of the Competition and Exercise Office of the China Cyberspace Security Association. His main research focuses on intelligent network attack and defense, industrial internet security, and artificial intelligence security. As principal investigator, he has led numerous national key research and development projects, grants from the National Natural Science Foundation of China, projects under the Guangdong Provincial Key Areas R&D Program, and security-specific projects for China Southern Power Grid. His research outcomes have been published in over 60 papers in domestic and international journals and conferences, alongside more than 40 patents. His awards and honors include the Guangdong Provincial Science and Technology Progress Award (First Prize), the Guangdong Provincial Philosophy and Social Sciences Outstanding Achievement Award (Second Prize), the Guangdong Provincial Outstanding Online Teaching Case Award during the pandemic phase, the Outstanding Case Award for Industry-Academia Collaboration in Cyberspace Security Education, the "Typical Practice Case in AI Security" from the China Cyberspace Security Association, and the China Industry-University-Research Cooperation Achievement Award (Second Prize).
He participated in the development of the "Shield Cube-Four Honeypots" active defense system. For major event cybersecurity assurance, the system ensured the platform security for the 24th Beijing Winter Olympics, the Hangzhou Asian Games, the Chengdu Universiade, the Harbino Asian Winter Games, and The 15th National Games of the People's Republic of China. It has also supported cybersecurity attack-defense exercises for the Ministry of Education and the Guangdong Provincial "Yuedun" exercises, receiving multiple letters of appreciation from key national departments.
He participated in the development of the "Shield Cube-Four Honeypots" active defense system. For major event cybersecurity assurance, the system ensured the platform security for the 24th Beijing Winter Olympics, the Hangzhou Asian Games, the Chengdu Universiade, the Harbino Asian Winter Games, and The 15th National Games of the People's Republic of China. It has also supported cybersecurity attack-defense exercises for the Ministry of Education and the Guangdong Provincial "Yuedun" exercises, receiving multiple letters of appreciation from key national departments.
Her main research interests include Data Security and Privacy Protection. She has been selected for the Guangdong Provincial Association for Science and Technology Young Science and Technology Talent Cultivation Program. She has published over 20 papers in top-tier international conferences and journals. She has led 5 research projects, including those funded by the National Natural Science Foundation of China and the China Postdoctoral Science Foundation. She has won the Second Prize of the Guangdong Provincial Science and Technology Progress Award, the ACM Qingdao Rising Star Award, and others.
Her research interests span Cyberspace Security, AI Agent Security, Software Security and Mobile Security. She has published more than 20 papers in leading international security conferences, including IEEE Symposium on Security and Privacy (IEEE S&P) and ACM Conference on Computer and Communications Security (ACM CCS). She has served as the principal investigator of multiple research projects funded by the National Natural Science Foundation of China, the National Key Research and Development Program of China, the Guangzhou Science and Technology Program, and the CCF-NSFOCUS Kunpeng Research Fund. Her research has also resulted in open-source foundational software for security analysis, which has been adopted by researchers and research groups at leading universities worldwide, including Purdue University, The Ohio State University, the University of Michigan, the University of Illinois Urbana-Champaign (UIUC), and the University of Southern California (USC).
Zefeng Lu is an Associate Professor in the School of Cyberspace Security at Guangzhou University. His research focuses on adversarial attacks and defenses for cross-modal foundation models and multimodal large language model security. He has published over 20 papers in leading international journals, including IEEE TIFS, TMM, TNNLS, and TITS. His honors and awards include the First Prize of the Guangdong Provincial Science and Technology Progress Award, a China Scholarship Council (CSC) Scholarship, a Silver Award at the World Intelligent Challenge, and the Tencent Founder Innovation Award. He has also received funding from the Young Scientists Fund of the National Natural Science Foundation of China (NSFC). He is a member of the China Society of Image and Graphics (CSIG) and serves as a reviewer for leading journals in information security, multimedia, machine learning, intelligent transportation, and computer vision.
Workshop 12
Submission Link for Workshop 12
Summary:
The rapid digitalization of industrial systems has generated large-scale spatio-temporal data across manufacturing, e-commerce, logistics, energy, transportation, and other industrial scenarios. This workshop focuses on spatio-temporal intelligence modeling for such complex environments, including representation learning, graph-based modeling, dynamic relation learning, state evolution modeling, spatio-temporal forecasting, multimodal data fusion, and cross-domain knowledge transfer. It aims to bring together researchers and practitioners to explore effective methods for capturing dynamic dependencies, heterogeneous structures, and evolving operational states, while promoting scalable, adaptive, and generalizable spatio-temporal intelligence for real-world industrial applications.
Keywords: Spatio-Temporal Intelligence; Industrial Scenarios; Spatio-Temporal Modeling; Graph Learning; Dynamic Relation Modeling; Spatio-Temporal Forecasting; Industrial Intelligence
Daxin Zhu is the Dean of the School of Computer Science, Quanzhou Normal University, Quanzhou, China. He has published over 60 academic papers in renowned international journals, such as IEEE Transactions on Image Processing and Theoretical Computer Science, with more than 40 of them indexed by SCI and EI. He has led or participated in more than 20 scientific research projects, including Fujian Provincial Natural Fund Project, Key Industry-Academia-Research Projects of Fujian Provincial Education Department, and Quanzhou High-Level Talent Program. He has long been engaged in research on industrial internet applications, artificial intelligence applications, and big data technologies.
Zhibei Wu received the Ph.D. degree in applied mathematics with the School of Mathematical Sciences, Tongji University, Shanghai, China. He is currently a Lecturer with the School of Mathematical and Computer Sciences, Quanzhou Normal University. His current interests are in vehicle platoon and adaptive control
Qiuling Chen received the Ph.D. degree from the College of Computer and Cyber Security at Fujian Normal University, Fuzhou, China. She is currently a Lecturer with the School of Mathematical and Computer Sciences, Quanzhou Normal University. Her research interests include blockchain, network security, and location privacy.
Xianwei Guo received the Ph.D. degree from the College of Computer and Data Science, Fuzhou University, in 2025. He is currently a Lecturer with the School of Computer and Data Science, Minjiang University. His current research interests include ubiquitous computing, mobile crowdsensing, and spatiotemporal data modeling.
Canyang Guo received the Ph.D. degree from the College of Computer and Data Science, Fuzhou University, in 2025. He is currently a Lecturer with the School of Mathematical and Computer Sciences, Quanzhou Normal University. He has published over 30 journal articles, conference articles, and patents. His research interests include deep learning, spectral graph theory, big data, intelligent transportation systems, and wireless positioning.