US 12,513,102 B2
Simulation of a user of a social networking system using a language model
Andrew Garrod Bosworth, San Mateo, CA (US)
Assigned to Meta Platforms Technologies, LLC, Menlo Park, CA (US)
Filed by Meta Platforms Technologies, LLC, Menlo Park, CA (US)
Filed on Nov. 29, 2023, as Appl. No. 18/523,545.
Prior Publication US 2025/0175448 A1, May 29, 2025
Int. Cl. H04L 51/52 (2022.01); G06F 40/40 (2020.01)
CPC H04L 51/52 (2022.05) [G06F 40/40 (2020.01)] 18 Claims
OG exemplary drawing
 
1. A method comprising:
receiving a language model trained using training data obtained from user interactions performed by a target user of a social networking system;
receiving a set of content items posted on the social networking system that are relevant to the target user, wherein a first content item of the set of content items is posted by a second user of the social networking system, the set of content items being obtained from a newsfeed provided to the target user by the social networking system;
identifying a second content item relevant to the target user from the set of content items based on a ranking of the second content item in relation to rankings of other content items of the set of content items;
transmitting a prompt to the language model to obtain an output from the language model, the prompt describing the second content item and contextual information associated with the second content, and requesting the language model to recommend a user interaction with the second content item that the target user would be predicted to have performed upon viewing the second content item on the newsfeed, the user interaction including one or both of an interaction with the second content item or a textual response to the second content item;
receiving the user interaction from the language model in accordance with transmitting the prompt to the large language model; and
interacting with the second content in accordance with the user interaction.