All of human cooking compressed into 2 megabytes
Recorded: May 27, 2026, 1:21 p.m.
| Original | Summarized |
[2605.22391] Epicure: Navigating the Emergent Geometry of Food Ingredient Embeddings
Skip to main content Learn about arXiv becoming an independent nonprofit. We gratefully acknowledge support from the Simons Foundation, member institutions, and all contributors. > cs > arXiv:2605.22391 Help | Advanced Search All fields Search GO quick links Login Computer Science > Artificial Intelligence arXiv:2605.22391 (cs) [Submitted on 21 May 2026] Abstract:We present Epicure, a family of three sibling skip-gram ingredient embeddings retrained from scratch on a multilingual recipe corpus. We aggregate 4.14M recipes from 11 sources spanning seven languages, English, Chinese, Russian, Vietnamese, Spanish, Turkish, Indonesian, German, and Indian-English, and normalise the raw ingredient strings to 1,790 canonical entries via an LLM-augmented pipeline. A 203,508-edge ingredient-ingredient NPMI graph and an 80,019-edge typed FlavorDB ingredient-compound graph, 2,247 typed compound nodes across 15 categories, seed three Metapath2Vec variants that share architecture and hyperparameters and differ only in the random-walk schema: Cooc walks the co-occurrence graph only, Chem walks the typed compound metapaths only, and Core blends both via injected ingredient-ingredient walks at controlled mixing, placing each model at a distinct point on the chemistry-vs-recipe-context spectrum. Subjects: Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Computers and Society (cs.CY) Cite as: Focus to learn more arXiv-issued DOI via DataCite (pending registration) Submission history From: Josef Liyanjun Chen [view email] [v1]
Full-text links: View a PDF of the paper titled Epicure: Navigating the Emergent Geometry of Food Ingredient Embeddings, by Jakub Radzikowski and Josef ChenView PDFHTML (experimental)TeX Source view license Current browse context: < prev | new Change to browse by: References & Citations NASA ADSGoogle Scholar export BibTeX citation BibTeX formatted citation loading... Data provided by: Bookmark
Bibliographic Tools Bibliographic and Citation Tools Bibliographic Explorer Toggle Bibliographic Explorer (What is the Explorer?) Connected Papers Toggle Connected Papers (What is Connected Papers?) Litmaps Toggle Litmaps (What is Litmaps?) scite.ai Toggle scite Smart Citations (What are Smart Citations?) Code, Data, Media Code, Data and Media Associated with this Article alphaXiv Toggle alphaXiv (What is alphaXiv?) Links to Code Toggle CatalyzeX Code Finder for Papers (What is CatalyzeX?) DagsHub Toggle DagsHub (What is DagsHub?) GotitPub Toggle Gotit.pub (What is GotitPub?) Huggingface Toggle Hugging Face (What is Huggingface?) ScienceCast Toggle ScienceCast (What is ScienceCast?) Demos Demos Replicate Toggle Replicate (What is Replicate?) Spaces Toggle Hugging Face Spaces (What is Spaces?) Spaces Toggle TXYZ.AI (What is TXYZ.AI?) Related Papers Recommenders and Search Tools Link to Influence Flower Influence Flower (What are Influence Flowers?) Core recommender toggle CORE Recommender (What is CORE?) Author About arXivLabs arXivLabs: experimental projects with community collaborators Which authors of this paper are endorsers? | About contact arXivClick here to contact arXiv subscribe to arXiv mailingsClick here to subscribe Copyright Web Accessibility Assistance arXiv Operational Status |
The work introduces Epicure, a family of three sibling skip-gram ingredient embeddings that are retrained from scratch using a multilingual recipe corpus. This foundation is built upon aggregating 4.14 million recipes drawn from eleven sources covering seven languages, including English, Chinese, Russian, Vietnamese, Spanish, Turkish, Indonesian, German, and Indian-English. To standardize the input, the raw ingredient strings were processed and normalized into 1,790 canonical entries through a pipeline augmented by a large language model. From this data, the authors constructed two primary knowledge graphs: an ingredient-ingredient NPMI graph containing 203,508 edges, and a typed FlavorDB ingredient-compound graph featuring 80,019 edges across 2,247 typed compound nodes categorized into 15 classes. To leverage these structures, the authors seeded three variants of the Metapath2Vec model, all sharing the same underlying architecture and hyperparameters but differing in their random-walk schemas. One variant, Cooc, traverses only the co-occurrence graph. Another, Chem, explores only the typed compound metapaths. The third variant, Core, fuses these approaches by incorporating injected ingredient-ingredient walks at controlled mixing points. This allows the models to be positioned distinctly along a spectrum ranging from chemistry principles to recipe context, offering a nuanced representation of food ingredient relationships. |