Epicure: Multidimensional Flavor Structure in Food Ingredient Embeddings
For computational gastronomy and food AI, this work provides a method to extract interpretable flavor dimensions from existing embeddings, but the novelty is incremental as it applies known techniques (LLM curation, embedding analysis) to a specific domain.
The authors show that FlavorGraph's ingredient embeddings encode tacit culinary knowledge (taste, texture, geography, etc.) and recover at least fifteen classifiable dimensions using an LLM-augmented curation pipeline that consolidates 6,653 ingredients into 1,032 canonical entries.
A chef's intuition about flavor, texture, and cultural identity represents tacit knowledge that is difficult to articulate yet central to culinary practice. We show that this knowledge is already encoded in FlavorGraph's 300-dimensional ingredient embeddings, trained on recipe cooccurrence and food chemistry, and that it can be systematically recovered. An LLM-augmented curation pipeline consolidates 6,653 raw FlavorGraph ingredients into 1,032 canonical entries, substantially strengthening the recoverable structure. We identify at least fifteen independently classifiable dimensions spanning taste, texture, geography, food processing, and culture.