US 11,995,394 B1
Language-guided document editing
Vlad Ion Morariu, Potomac, MD (US); Puneet Mathur, College Park, MD (US); Rajiv Bhawanji Jain, Falls Church, VA (US); Jiuxiang Gu, Baltimore, MD (US); and Franck Dernoncourt, Spokane, WA (US)
Assigned to ADOBE INC., San Jose, CA (US)
Filed by ADOBE INC., San Jose, CA (US)
Filed on Feb. 7, 2023, as Appl. No. 18/165,579.
Int. Cl. G06F 40/166 (2020.01); G06F 3/16 (2006.01); G06F 40/284 (2020.01); G06N 20/00 (2019.01); G10L 15/22 (2006.01); G10L 15/26 (2006.01)
CPC G06F 40/166 (2020.01) [G06F 3/167 (2013.01); G06F 40/284 (2020.01); G06N 20/00 (2019.01); G10L 15/22 (2013.01); G10L 15/26 (2013.01)] 19 Claims
OG exemplary drawing
 
1. A method comprising:
obtaining a document and a natural language edit request;
generating a structured edit command using a machine learning model based on the document and the natural language edit request by performing object detection on the document to obtain a text object, performing text recognition on the document to obtain a text embedding; and combining the text object with the text embedding to obtain a text-enriched object, wherein the structured edit command is generated based on the text-enriched object; and
generating a modified document based on the document and the structured edit command, wherein the modified document comprises a revision of the document that incorporates the natural language edit request.