Food and Health Knowledge Graphs Using Large Language Models (LLMs)

About the Workshop

Substantial emphasis has traditionally been placed on agricultural techniques to increase production and studying how nutrition affects human health—yet the relationships across the farm-to-fork continuum remain difficult to represent, harmonize, and evaluate in a way that supports health informatics use cases.

In parallel, Knowledge Graphs (KGs) and Large Language Models (LLMs) are rapidly converging. This workshop focuses on Food & Health KGs built with LLMs and LLM-aided methods, including ontology/terminology modeling, semi-automated KG construction, alignment with clinical standards, and hybrid KG–LLM architectures (e.g., RAG).

Goals

• Share methods, systems, and datasets for LLM-aided Food & Health KG pipelines.
• Establish reusable design patterns for KG–LLM integration in health informatics.
• Produce community artifacts: An evaluation checklist/rubric and a perspective article outline.

Call for Papers

We invite submissions related to the convergence of semantic technologies and generative AI in the food and health domain. Topics of interest include (but are not limited to):

• LLM-assisted extraction/linking for food, nutrition, supplements, and exposures.
• Ontology-guided Knowledge Graph (KG) construction and semantic validation.
• Hybrid KG–LLM systems (Retrieval-Augmented Generation [RAG] over curated KGs).
• Integration and alignment across food/nutrition resources and public health ecosystems.
• Evaluation frameworks for LLM-aided KG quality and trustworthiness.
• Responsible AI: Privacy, consent, fairness, transparency, and governance.

Submission Types

All submissions must follow the IEEE ICHI 2026 format.

• Full papers: Up to 10 pages.
• Short papers: Up to 6 pages.
• Extended abstracts/posters: Up to 2 pages.