Best Paper Award at the RAGE-KG 2025 Workshop for Wimmics members

Yousouf Taghzouti, Franck Michel, Tao Jiang, Louis-Felix Nothias, and Fabien Gandon received the Best Paper Award for “User Interface and Agent Interface for Online Generation of Knowledge Graph’s Competency Questions and Question-Query Training Sets” at the RAGE-KG 2025 Workshop, held in Nara, Japan, on November 3.

An increasing number of domains and professions are seeking to leverage the latest generative AI techniques to facilitate access to their databases. However, there are still very few datasets of question–query pairs that enable the fine-tuning of large language models for tasks such as translating natural language questions into queries over graph databases (SPARQL).

This paper introduces Q2Forge, a web application designed to facilitate the creation of question–query pairs for any RDF knowledge graph. The tool allows users to generate, test, and refine competency questions and their SPARQL equivalents directly within the interface. It relies on a retrieval-augmented generation (RAG) architecture for contextual enrichment. The result is an open-source solution for creating reusable question–query datasets, applicable to any knowledge graph.

Louis-Felix Nothias (Institut de Chimie de Nice) and Fabien Gandon (Inria Wimmics, CNRS, i3S) are researchers involved in the KG-Bot project, supported by the RISE Academy. Within this framework, the KG-Bot project contributed to the development of a proof of concept through preparatory work that laid the foundation for the launch of the MetaboLinkAI project. Both researchers are now members of this international research project, funded by the French National Research Agency (ANR) and the Swiss National Science Foundation (SNSF) for the 2025–2029 period.

To read the paper: https://hal.science/hal-05289962