Foundation Models and Agentic Intelligence for Graph-Centric Applications

Scope Graph-structured data play a fundamental role in modeling complex relations across social networks, recommender systems, knowledge graphs, cybersecurity, biomedicine, scientific discovery, urban computing, and enterprise intelligence. Recent advances in foundation models, large language models, and agentic intelligence are creating new opportunities for graph-centric applications by enabling large-scale pre-training, cross-domain transfer, prompt-based adaptation, semantic understanding, graph reasoning, retrieval, tool use, planning, and interactive decision-making over graph-structured data. Their convergence is opening a timely research frontier that connects graph learning, language-based intelligence, and autonomous agent systems. This aims to provide a focused venue for high-quality research on Foundation Models and Agentic Intelligence for Graph-Centric Applications. It welcomes theoretical, methodological, empirical, and system-oriented contributions that advance both directions of this emerging intersection: foundation models and agents for graph learning, reasoning, and decision-making, as well as graph-enhanced methods for improving language models, retrieval-augmented generation, autonomous agents, and knowledge-intensive AI systems. Particular emphasis will be placed on graph foundation models, graph-enhanced language models, graph-based reasoning and retrieval, graph agents, trustworthy and scalable graph-centric applications, benchmarks and evaluation protocols, and real-world applications in complex relational domains. Topics of Interest We welcome original research papers that address, but are not limited to: 1. Graph Foundation Models and Graph-Centric Pre-training Large-scale graph pre-training and graph foundation models Self-supervised, weakly supervised, and instruction-based graph learning Graph prompt learning, graph tuning, and parameter-efficient adaptation Transfer learning, domain adaptation, and continual learning on graphs Dynamic, temporal, heterogeneous, and multi-domain graph foundation models 2. Foundation Models for Graph Learning and Reasoning Large language models for graph learning, graph analytics, and graph reasoning Multimodal foundation models for graph-structured data Language-enhanced graph representation learning Foundation-model-based graph mining, prediction, and decision-making Neural, symbolic, and neuro-symbolic graph reasoning 3. Graphs for Large Language Models and Agentic Intelligence Graph-enhanced large language models and foundation models Knowledge graphs and structured knowledge for language model reasoning Graph-based context modeling, memory construction, and long-context organization Path-aware, subgraph-aware, and relation-aware reasoning with LLMs Graph-based factuality enhancement, alignment, and evaluation of language models 4. Graph Retrieval-Augmented Generation and Graph Agents Graph retrieval-augmented generation and graph-based evidence retrieval Graph agents and autonomous agents over graph-structured environments Graph-based planning, tool use, and interactive decision-making Graph world models and executable knowledge graphs for agents Multi-agent collaboration and agentic workflows over graph-structured data 5. Trustworthy, Scalable, and Deployable Graph-Centric Systems Robust, fair, explainable, and privacy-preserving graph foundation models Security, safety, and alignment of graph-enhanced AI systems Scalable training, inference, compression, and deployment Efficient graph learning systems and open-source platforms Benchmarks, datasets, evaluation protocols, and reproducibility studies 6. Foundation Models and Agentic Intelligence for Graph-Centric Applications Recommender systems, social networks, and web intelligence Cybersecurity, fraud detection, and risk analysis Biomedicine, healthcare, and scientific discovery Urban computing, transportation, and spatiotemporal intelligence Finance, enterprise intelligence, and complex decision-support systems Types of Submissions We welcome original and high-quality contributions in the following categories types: Original research papers Survey and review papers System and benchmark papers Application and industrial papers Academic Editors Prof. Qingyun Sun (sunqy@buaa.edu.cn), School of Computer Science and Engineering, Beihang University, Beijing, China. Prof. Ziwei Zhang (zwzhang@buaa.edu.cn), School of Computer Science and Engineering, Beihang University, Beijing, China. Dr. Haonan Yuan (yuanhn@buaa.edu.cn), School of Computer Science and Engineering, Beihang University, Beijing, China. Prof. Jianxin Li (lijx@buaa.edu.cn), School of Computer Science and Engineering, Beihang University, Beijing, China. Prof. Philip S. Yu (psyu@uic.edu), Department of Computer Science‌, University of Illinois at Chicago, Chicago, United States. Submission Guidelines To submit your manuscript, please go to the Transactions on Graph Intelligence and Network Applications journal website at Add Submissions - Scilight Press and follow the procedures for manuscript submission. Author Submission Guidelines can be found at: Instruction for Authors - | Transactions on Graph Intelligence and Network Applications. Please make sure to include the title of this Call for Papers in the cover letter. There is no Article Processing Charge for all submissions and accepted papers. Important Dates Submission Deadline: 31 August 2026 First Round Review Decision: 31 September 2026 Final Decision: 30 October 2026 Expected Publication: 15 November 2026 All deadlines are end-of-day in the Anywhere on Earth (AoE) time zone. All accepted papers will be published in the earliest available issue. The above timeline is provided for reference only, and the actual schedule may be subject to change. All manuscripts will be peer-reviewed in accordance with the journal's established policies and procedures. The final papers will be selected for publication depending on the results of the peer-review process and the reviews of the Academic Editors and Editor-in-Chief.