Special Presentation by Mahdi Boloursaz Mashhadi (U. of Surrey, UK)

Hosted by the Future Networks Artificial Intelligence & Machine Learning (AIML) Working Group

Date/Time: Thursday, 21 May 2026 @ 12:00 UTC (1 PM BST)

Topic:

6G-Agents: From LLM-Driven Interpretability to Autonomous Agentic AI for 6G and Beyond 

Abstract:

The evolution of 6G networks towards software-defined, AI-native, and multi-vendor architectures introduces unprecedented complexity in network management, security, and interoperability. While recent advances in large language and multimodal models (LLMs/MLLMs) offer powerful capabilities for reasoning, interpretation, and automation, their role in telecom systems still remains reactive with human-in-the-loop. This talk presents a unified vision that transitions from LLM-driven human interpretability frameworks to fully agentic AI systems for autonomous network operation. We introduce agentic AI architectures that extend LLMs from passive reasoning engines to active decision-makers through perception–planning–action loops, augmented by retrieval-augmented generation (RAG) and multi-agent coordination. These systems enable automated intent-driven network control, self-healing and troubleshooting, and compliance enforcement frameworks. For autonomous security compliance, an agentic AI workflow is demonstrated which continuously monitors, evaluates, and enforces evolving standards (e.g., 3GPP, O-RAN) using knowledge-grounded reasoning and iterative refinement loops. Experimental insights highlight the trade-offs between performance, interpretability, latency, and implementation costs. This talk introduces agentic AI as a major enabler for transforming future networks to fully autonomous, intent-driven, and secure infrastructures with human interpretability.

Speaker:

Mahdi Boloursaz Mashhadi (Senior Member, IEEE) is a Lecturer at the 5G/6G Innovation Centre (5G/6GIC) at the Institute for Communication Systems (ICS), University of Surrey (UoS), UK, and a Surrey AI fellow. His research is focused on the intersection of AI/ML with wireless communication, learning and communication co-design, generative AI for telecommunications, Token Communications (TokenCom), and collaborative machine learning. Prior to joining ICS, he was a postdoctoral research associate at the Intelligent Systems and Networks (ISN) Research Group, Imperial College London, 2019-2021. He received the B.S., M.S., and Ph.D. degrees in mobile telecommunications from the Sharif University of Technology (SUT), Tehran, Iran. He has (co-/)organized and (co-/)chaired several workshops, tutorials, and special sessions at the intersection of AI/ML and wireless communications in various IEEE venues including ICC, ICMLCN, SPAWC 2025, and GLOBECOM. He was a PI/Co-PI for various government and industry funded projects including the DSIT 12M£ UK national project TUDOR. He is the recipient of the Best Paper Award from several IEEE conferences/workshops including INFOCOM, EWDTS, and the Exemplary Reviewer Award from the IEEE ComSoc in 2021-2023. He served as a panel judge for the International Telecommunication Union (ITU) on the “AI/ML in 5G” challenge 2021-2022. He is an associate editor for the Springer Nature Wireless Personal Communications.

Brochure (PDF): 6G-Agents: From LLM-Driven Interpretability to Autonomous Agentic AI for 6G and Beyond