US 12,229,525 B2
Controllable reading guides and natural language generation
Barak Peleg, Givatayim (IL); Dan Padnos, Tel Aviv (IL); Amnon Morag, Tel Aviv (IL); Gilad Lumbroso, Kfar Saba (IL); Yoav Shoham, Tel Aviv (IL); Ori Goshen, Tel Aviv-Jaffa (IL); Barak Lenz, Tel Aviv (IL); Or Dagan, Tel Aviv (IL); and Guy Einy, Tel Aviv (IL)
Assigned to AI21 LABS, Tel Aviv (IL)
Filed by AI21 LABS, Tel Aviv (IL)
Filed on May 11, 2023, as Appl. No. 18/315,654.
Application 18/315,654 is a division of application No. 18/153,610, filed on Jan. 12, 2023.
Application 18/153,610 is a continuation of application No. PCT/US2021/041428, filed on Jul. 13, 2021.
Claims priority of provisional application 63/187,162, filed on May 11, 2021.
Claims priority of provisional application 63/187,170, filed on May 11, 2021.
Claims priority of provisional application 63/086,254, filed on Oct. 1, 2020.
Claims priority of provisional application 63/084,500, filed on Sep. 28, 2020.
Claims priority of provisional application 63/051,288, filed on Jul. 13, 2020.
Prior Publication US 2023/0281398 A1, Sep. 7, 2023
Int. Cl. G06F 40/30 (2020.01); G06F 3/0482 (2013.01); G06F 40/166 (2020.01); G06F 40/211 (2020.01); G06F 40/274 (2020.01); G06F 40/289 (2020.01); G06F 40/56 (2020.01); G06F 40/58 (2020.01); G06F 3/0486 (2013.01)
CPC G06F 40/56 (2020.01) [G06F 3/0482 (2013.01); G06F 40/166 (2020.01); G06F 40/211 (2020.01); G06F 40/274 (2020.01); G06F 40/289 (2020.01); G06F 40/30 (2020.01); G06F 40/58 (2020.01); G06F 3/0486 (2013.01)] 11 Claims
OG exemplary drawing
 
1. A non-transitory computer readable medium including instructions that when executed by one or more processing devices cause the one or more processing devices to perform a method including:
receiving from a user an identification of a plurality of different text files;
analyzing text from each of the plurality of different text files;
identifying concepts conveyed by the text from each of the plurality of different text files;
determining an ordering for the identified concepts to be used in generating an output text;
receiving from the user a selection of at least one of a template or an example document;
generating, using one or more machine learning models, the output text based on the determined ordering for the identified concepts and the received selection, wherein the generated output text conveys each of the identified concepts and includes one or more text elements not included in the text of the plurality of different text files;
inserting the generated output text into a document based on the received selection without further input from the user;
receiving from the user an identification of a location in the generated output text in the document for at least one text revision;
receiving text input entered by a user in a field of a graphical user interface configured to display the generated output text;
automatically generating, using the one or more machine learning models, one or more text revision options, based on a context of the generated output text before or after the identified location and also based on a meaning associated with the text input received from the user, and causing the one or more text revision options to be displayed to the user;
receiving, from the user, a selection of a text revision option from among the one or more text revision options;
updating the generated output text by causing the selected text revision option to be included in the generated output text at a location that includes the identified location; and
causing the updated output text to be shown on the graphical user interface.