Competition

Start: June 15, 2025, midnight

Description: The goal is to automatically extract knowledge jointly from tables and related texts. For example, in the figure below, a participant tool should extract triples composed of mentions located partly in the text and partly in the table and disambiguated with entities and predicates identified in the Wikidata reference knowledge graph. Knowledge extraction will be carried out in Open Extraction Information mode, using the Wikidata knowledge graph as a reference, but with the possibility that certain predicates or entities may not be present in the graph. Similar to the Text2KGBench evaluation, and because the set of triples are not exhaustive for a given sentence, to avoid false negatives, we will follow a locally closed approach by only considering the relations that are part of the ground truth. The evaluation will then use metrics such as P, R and F1.