Epicure: Navigating the Emergent Geometry of Food Ingredient Embeddings
Researchers have developed "Epicure," a set of three skip-gram embeddings trained on a large multilingual recipe corpus. These embeddings are designed to capture the relationships between food ingredients, considering both co-occurrence in recipes and chemical compound data. The models, named Cooc, Chem, and Core, offer different balances between recipe context and chemical properties, providing a nuanced understanding of ingredient interactions. AI
IMPACT Introduces novel embeddings for food ingredients, potentially enabling new applications in recipe generation and food science.