The rapid evolution of Large Language Models (LLMs) is reshaping the landscape of knowledge base and database management, creating new opportunities as well as challenges. LLMs are opening up novel pathways for core knowledge base and database tasks, including data cleansing, schema and relation inference, knob tuning, and query processing, while knowledge bases and database systems are increasingly being leveraged to enhance LLM reliability, mitigate hallucination, and improve data governance and privacy. Despite growing interest and early successes in both LLM-empowered knowledge base/database systems and knowledge base/database-enhanced LLMs, this research area remains at an early and dynamic stage. We invite submissions that explore the synergy, integration, and co-evolution of knowledge bases, database systems, and LLMs, and that provide insights, solutions, or emerging applications from academic and industrial perspectives.

The workshop welcomes theory and methodology papers falling into the scope of following themes, including but not limited to:

• LLM-empowered Knowledge and Data Engineering
  • LLM-enabled Improvements to Foundational Knowledge Base and Database Algorithms
  • Database Design, Configuration, and Tuning with LLMs
  • Data Cleansing and Augmentation with LLMs
  • Data Indexing and Query Optimization with LLMs
  • Reasoning Over Knowledge Bases with LLMs
  • Automated Data Understanding and Visualization with LLMs
  • Novel Query Interfaces and Interactive Query Refinement with LLMs
  • Context-aware Data Retrieval with LLMs
  • Personalized Query Optimization Using LLMs


• Knowledge Base and Database-enhanced LLMs
  • Data Collection and Preparation for LLMs
  • Database-inspired Techniques for Modeling, Storage, and Provenance of LLMs
  • Vector Data Management for LLMs
  • Data and Metadata Management for LLM Life Cycle
  • Data Management for Multi-modal LLMs
  • Retrieval-augmented Generation and Reasoning for LLMs
  • Novel Data Management Systems for Accelerating LLM Training

Papers should not exceed 10 pages (including references). No appendix is allowed. Manuscripts must be self-contained, in English, and in PDF format according to the IEEE format available at https://www.ieee.org/conferences/publishing/templates, using the generic "conference" template. The PDF files must have all non-standard fonts embedded. Submissions will be reviewed double-blind, and author names and affiliations should NOT be listed. Submitted works will be assessed based on their originality, impact, technical quality, relevance, and clarity. Please refer to the ICDE'26 website for further details.

Note that at least one author of each accepted paper must register for the workshop (details to come on the ICDE'26 website). For questions about submission, please contact us at: kdexllm2026@gmail.com

Papers should be submitted using the Conference Management Tool https://cmt3.research.microsoft.com/KDExLLM2026.