Methods and systems for parsing a mix of features and instructions into a prompt
AssigneeShopify Inc.
InventorDavid Goligorsky
Life of this patent
Estimating the term…
Abstract
A computer system and computer-implemented method, the method including receiving text input, the text input including feature inputs and prompt instructions; analyzing the text input to identify the feature inputs and the prompt instructions; and generating a prompt to be provided to a Large Language Model (LLM) based on a prompt template, the feature inputs, and the prompt instructions.
Claims20
A computer-implemented method for parsing a mix of features inputs and prompt instructions for a prompt, the method comprising:
- providing a user interface for receiving text input, the user interface providing for delineation between feature inputs describing content for a Large Language Model (LLM) output and prompt instructions describing a style for the LLM output;
- receiving the text input using the user interface, the text input including the feature inputs and the prompt instructions, the prompt instructions being delineated from the feature inputs by one or more tags;
- analyzing the text input to identify the feature inputs and the prompt instructions, including parsing the text input to identify the prompt instructions based on the one or more tags;
- generating the prompt to be provided to an LLM based on a prompt template, the feature inputs, and the prompt instructions;
- sending the prompt to the LLM;
- and
- receiving the LLM output from the LLM based on the prompt.
A computer system configured for parsing a mix of features inputs and prompt instructions for a prompt, the computer system comprising:
- one or more processors;
- a user interface;
- and
- a computer readable medium storing instructions that when executed by at least one of the one or more processors, cause the computer system to:
- provide for receiving text input at the user interface by providing for delineation between feature inputs describing content for a Large Language Model (LLM) output and prompt instructions describing a style for the LLM output;
- receive the text input, the text input including the feature inputs and the prompt instructions, the prompt instructions being delineated from the feature inputs by one or more tags;
- analyze the text input to identify the feature inputs and the prompt instructions, including parsing the text input to identify the prompt instructions based on the one or more tags;
- generate the prompt to be provided to an LLM based on a prompt template, the feature inputs, and the prompt instructions;
- send the prompt to the LLM;
- and
- receive the LLM output from the LLM based on the prompt.
A non-transitory computer readable storage medium storing instructions that, when executed by at least one processor of a computer system, cause the computer system to:
- provide a user interface at the computer system for receiving text input at the user interface by providing for delineation between feature inputs describing content for a Large Language Model (LLM) output and prompt instructions describing a style for the LLM output;
- receive the text input, the text input including the feature inputs and the prompt instructions, the prompt instructions being delineated from the feature inputs by one or more tags;
- analyze the text input to identify the feature inputs and the prompt instructions, including parsing the text input to identify the prompt instructions based on the one or more tags;
- generate the prompt to be provided to an LLM based on a prompt template, the feature inputs, and the prompt instructions;
- send the prompt to the LLM;
- and
- receive the LLM output from the LLM based on the prompt.
Dependencies are inferred from the claim text.