Back to results
US 2026/0289626 A1Published application
System
AssigneeSoftBank Group Corp.
InventorJunya NOMURA
Life of this patent
Estimating the term…
Abstract
A system includes a processor that is configured to receive, as input from a user, budget information, disliked ingredient information, and allergen information to be avoided, analyze the received information and, based on a result of the analysis, generate a prompt sentence that instructs a generative artificial intelligence model to generate a dinner recipe, and identify ingredients required for the generated recipe and prepare an order-ready state of the ingredients through an electronic commerce platform.
Claims20
A system comprising:
- circuitry configured to:
- receive, via a communication interface coupled to a packet-switched network, constraint parameters including at least a numerical constraint parameter, a first exclusion parameter, and a second exclusion parameter from a terminal device, and generate a structured condition incorporating the constraint parameters stored in a storage device;
- tokenize the structured condition, configure inference parameters including temperature and maximum output length, transmit the tokenized structured condition to an inference engine executing a generative AI model, and decode an output token sequence from the inference engine to obtain a structured output plan as a text string;
- and
- parse the text string to extract record identifiers and attribute descriptions, normalize the extracted identifiers to canonical record identifiers, and transmit the structured output plan to the terminal device via the communication interface.
2 more elements
A system comprising:
- circuitry configured to:
- receive, via a communication interface coupled to a packet-switched network, constraint parameters including a numerical constraint parameter, a first exclusion parameter, and a second exclusion parameter from a terminal device, and generate a structured condition incorporating the constraint parameters stored in a storage device;
- tokenize the structured condition, configure inference parameters, transmit the tokenized structured condition to an inference engine executing a generative AI model, and decode an output token sequence to obtain a structured output plan as a text string;
- parse the text string to extract record identifiers and attribute descriptions, normalize the extracted identifiers to canonical record identifiers, and apply a scoring function combining a constraint satisfaction score and a preference match score to rank the structured output plan;
- and
- transmit the ranked structured output plan to the terminal device via the communication interface, and update a preference model stored in the storage device based on evaluation information received from the terminal device.
3 more elements
A method comprising:
- receiving, via a communication interface coupled to a packet-switched network, constraint parameters including at least a numerical constraint parameter, a first exclusion parameter, and a second exclusion parameter from a terminal device, and generating a structured condition incorporating the constraint parameters stored in a storage device;
- tokenizing the structured condition, configuring inference parameters including temperature and maximum output length, transmitting the tokenized structured condition to an inference engine executing a generative AI model, and decoding an output token sequence from the inference engine to obtain a structured output plan as a text string;
- and
- parsing the text string to extract record identifiers and attribute descriptions, normalizing the extracted identifiers to canonical record identifiers, and transmitting the structured output plan to the terminal device via the communication interface.
1 more element
Dependencies are inferred from the claim text.