Abstract
Online recipe searches performed by users are often highly personalized; users seek specific culinary guidance, cooking methods, styles, etc. A large language model (LLM) can be used to generate recipe insights. However, an LLM can generate ungrounded results, and/or miss specific nuances of recipes along dimensions specified by the user. This disclosure describes techniques that leverage a large language model (LLM) to mine, aggregate, and summarize recipe information across multiple pages/sources and to deliver personalized, contextual, and insightful responses to user queries for recipes via a deep understanding of the pages/sources of the recipe. Recipe tips and guidance snippets are utilized to ground the LLM response to the context of a specific recipe. Endorsements and justifications are provided based on phrases or snippets extracted from recipe sources, including user generated content. The described techniques can capture a wider diversity of perspectives across web results. By grounding the LLM responses, the techniques enable traceability of the response back to verifiable sources and ensure their accuracy for tail queries.
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.
Recommended Citation
Kota, Nagaraj; Kukreja, Varsha; and Jindal, Jatin, "Deep Content Understanding for Recipe Guidance using a Large Language Model", Technical Disclosure Commons, (November 18, 2024)
https://www.tdcommons.org/dpubs_series/7550