Systems and methods for computer generation of a modifiable product description

A system comprising:

• at least one processor;
• and
• a memory storing processor-executable instructions that, when executed by the at least one processor, cause the system to:
• prompt a generative machine-learning (ML) model to cause the generative ML model to generate a text description;
• for a logits vector generated by the generative ML model:
• select an output token corresponding to a first entry of the logits vector to form a basis of a word or phrase in a segment of the text description;
• and
• store an alternative token corresponding to a second entry of the logits vector, the alternative token forming the basis of an alternative word or alternative phrase that is not used in the segment of the text description;
• and
• provide content for presentation, the content including both the text description and the alternative word or alternative phrase that may be substituted in place of the word or phrase in the segment of the text description.

A non-transitory computer-readable storage medium having stored thereon computer-executable instructions that, when executed by a computer, cause the computer to perform operations comprising:

• prompting a generative machine-learning (ML) model to cause the generative ML model to generate a text description;
• for a logits vector generated by the generative ML model:
• selecting an output token corresponding to a first entry of the logits vector to form a basis of a word or phrase in a segment of the text description;
• and
• storing an alternative token corresponding to a second entry of the logits vector, the alternative token forming the basis of an alternative word or alternative phrase that is not used in the segment of the text description;
• and
• providing content for presentation, the content including both the text description and the alternative word or alternative phrase that may be substituted in place of the word or phrase in the segment of the text description.