Archimedes NLP Theme Meeting: Invited Talk β Arijit Khan (Bowling Green State University)
Graph-native LLMs and Agentic Systems: Rethinking the AI Harness
Speaker: Arijit Khan, Associate Professor, Bowling Green State University (Ohio, USA)
Date and time: Monday 2 November 2026, 14:30β16:00 (Greek time)
Venue: Amphitheater, Archimedes Unit, Artemidos 1, 15125 Marousi, Attica, and online via Microsoft Teams.
Join the meeting on Microsoft Teams
Meeting ID: 317 060 152 094 132 9
Passcode: Ku6Hs3qz
Abstract
Large Language Models (LLMs) continue to advance rapidly, yet their limitations in structured and multi-hop reasoning highlight the need for graph-native AI systems, and for rethinking the AI harness: the scaffolding of retrieval, tools, memory, and orchestration that surrounds a model and largely determines how reliably it performs in practice. In the first part of this talk, I will present our recent progress on knowledge graph-based retrieval augmented generation (KG-RAG), which serves as a graph-native retrieval layer within the harness and improves LLM robustness across data science tasks such as domain-specific schema matching, entity alignment, and fact verification. In the second part, I will introduce our latest work on agentic LLM systems whose harness is itself graph-structured, spanning graphs of tools, actions, memory, and knowledge, and a new evaluation framework that identifies, measures, and explains where misalignment arises during graph-structured reasoning within the harness.
Speaker Bio
Arijit Khan is an Associate Professor at Bowling Green State University (Ohio, USA). His PhD is from the University of California, Santa Barbara, USA, and he did a postdoc in the Systems group at ETH Zurich, Switzerland. He has been an assistant professor at Nanyang Technological University, Singapore, and an associate professor at Aalborg University, Denmark. His research is on data management and AI for emerging problems in large graphs. He is an IEEE senior member and an ACM distinguished speaker.
Arijit is the recipient of the IBM Ph.D. Fellowship (2012β13), a VLDB Distinguished Reviewer award (2022), three SIGMOD Distinguished PC awards (2024, 2025, 2026), and a KDD Outstanding PC award (2025). He is the author of a book on uncertain graphs and over 130 publications in top venues including ACM SIGMOD, VLDB, IEEE TKDE, IEEE ICDE, ICLR, KDD, EMNLP, IJCAI, SIAM SDM, USENIX ATC, EDBT, The Web Conference (WWW), ACM WSDM, ACM CIKM, ACM TKDD, VLDB Journal, and ACM SIGMOD Record.
Dr Khan served as an associate editor of IEEE TKDE (2019β2024) and has served as an associate editor of ACM TKDD since 2023 and co-Editor-in-Chief of Knowledge Engineering Review since 2026. He also served as ACM CIKM short paper track co-chair (2024), IEEE ICDE demonstration paper track program co-chair (2025), ACM KDD PhD Consortium Track program co-chair (2025), and Australasian Database Conference (ADC) PC co-chair (2025).
Stay Tuned
For ways to receive news about the Archimedes NLP Theme and its meetings, and for the latest meeting information, visit the Archimedes NLP Theme calendar.
