Asst. Prof. Tome Eftimov delivered an invited talk at the INFOTECH-DATA Kick-Off Meeting in Valencia, presenting recent advances in AI-driven food knowledge systems, including FoodyLLM, FoodBench-QA 2026, and ontology-aware retrieval pipelines. The talk emphasized the importance of interoperable semantic resources, transparent benchmarking, and domain-specialized AI for transforming fragmented food data into actionable knowledge infrastructures.

Researchers from the Jožef Stefan Institute presented the FoodBench-QA shared task at the CL4Health Workshop collocated with LREC-COLING 2026, showcasing advances in trustworthy and semantically grounded AI for food and nutrition question answering.

Tome Eftimov presented 12+ years of research on NLP for food and nutrition at the EuroFIR Food Forum 2026, showcasing advances in transforming unstructured food data into interoperable knowledge through information extraction, ontologies, and AI (including FoodyLLM and FoodBench-QA).

At the DSI 2026 conference in Portorož, Matevž Ogrinc presented NutriBase, a web platform developed at the Jožef Stefan Institute for managing and integrating food composition data following FAIR principles. The platform connects national and international food databases, integrates with applications such as Veš, kaj ješ and Piskr, and uses AI methods and citizen contributions to continuously enrich the food data ecosystem.