US 12,147,775 B2
Content augmentation with machine generated content to meet content gaps during interaction with target entities
Niyati Himanshu Chhaya, Telangana (IN); Udit Kalani, Uttarakhand (IN); Roodram Paneri, Rajasthan (IN); Sreekanth Reddy, Andhra Pradesh (IN); Niranjan Kumbi, Fremont, CA (US); Navita Goyal, Bangalore (IN); Balaji Vasan Srinivasan, Bangalore (IN); and Ayush Agarwal, Rajasthan (IN)
Assigned to Adobe Inc., San Jose, CA (US)
Filed by Adobe Inc., San Jose, CA (US)
Filed on Oct. 14, 2021, as Appl. No. 17/501,602.
Prior Publication US 2023/0121711 A1, Apr. 20, 2023
This patent is subject to a terminal disclaimer.
Int. Cl. G06F 40/40 (2020.01); G06F 40/216 (2020.01); G06F 40/279 (2020.01); G06F 40/30 (2020.01); G06F 40/44 (2020.01); G06F 40/56 (2020.01); G06N 3/08 (2023.01); G06N 3/09 (2023.01)
CPC G06F 40/40 (2020.01) [G06F 40/216 (2020.01); G06F 40/279 (2020.01); G06F 40/30 (2020.01); G06F 40/44 (2020.01); G06F 40/56 (2020.01); G06N 3/08 (2013.01); G06N 3/09 (2023.01)] 17 Claims
OG exemplary drawing
 
1. A computer-implemented method for using generative transformer networks to generate output content, wherein the method includes performing, with one or more processing devices, operations comprising:
receiving a request to generate content for a target entity, wherein the request includes one or more keywords;
retrieving, for the target entity, a current stage identifier linking the target entity to a current stage within a multi-stage objective;
generating an input vector including the current stage identifier, a target stage identifier, a token embedding comprising the one or more keywords, and a position embedding for each of the one or more keywords, the target stage identifier associated with a target stage within the multi-stage objective different from the current stage;
determining causally significant features of output text using a support vector machine model trained to determine the causally significant features using an average treatment effect approach;
generating output text content for the target entity by applying a generative transformer network to the input vector, wherein the generative transformer network is trained to determine the output text predicted to cause a transition of the target entity from the current stage to the target stage; and
transmitting the output text content to a computing device associated with the target entity.