Large language models (LLMs) make it possible to build knowledge graphs from professional literature at scale, but how such graphs fail at that scale is rarely measured. We built \sys{}, a cross-media knowledge graph of swine, aquaculture, tomato, maize and rice production, and analysed its 199,418 entities and 277,053 relations together with the web system that serves it. An LLM extracted typed triples from the professional literature of each sector; 74.9% of the relations carry an evidence record of document, page and excerpt, and a multimodal embedding model links 25,158 entities to images and 14,132 to video clips. We identified five failure modes of construction. First, the sector schem... More >

Expert advice on crop management is often written informally, and in such text a number can easily be assigned the wrong role or attached to the wrong operation, material, or measurement basis. We propose QS-IE, a knowledge extraction method constrained by quantity and operation scope. Its unit of extraction is a complete quantitative fact. Given a shared set of numeric candidates and a single base model, QS-IE selects complete candidates from several prompting strategies, verifies them against the source text, and applies scope constraints, so that every extracted quantity stays linked to its source evidence. We built a corpus of 4006 texts on tomato cultivation. Two researchers with agrono... More >