US 11,869,506 B2
Selectively generating expanded responses that guide continuance of a human-to-computer dialog
Michael Fink, Tel Aviv (IL); Vladimir Vuskovic, Zollikerberg (CH); Shimon Or Salant, Rehovot Center District (IL); Deborah Cohen, Tel Aviv (IL); Asaf Revach, Nordia Center District (IL); David Kogan, Natick, MA (US); Andrew Callahan, Somerville, MA (US); Richard Borovoy, Boston, MA (US); Andrew Richardson, Cambridge, MA (US); Eran Ofek, Rehovot (IL); Idan Szpektor, Kfar Saba (IL); Jonathan Berant, Mountain View, CA (US); and Yossi Matias, Tel Aviv (IL)
Assigned to GOOGLE LLC, Mountain View, CA (US)
Filed by Google LLC, Mountain View, CA (US)
Filed on Jan. 28, 2022, as Appl. No. 17/587,478.
Application 17/587,478 is a continuation of application No. 16/621,376, granted, now 11,238,864, previously published as PCT/US2019/034625, filed on May 30, 2019.
Claims priority of provisional application 62/679,919, filed on Jun. 3, 2018.
Prior Publication US 2022/0157308 A1, May 19, 2022
Int. Cl. G06F 40/35 (2020.01); G06N 3/0475 (2023.01); G10L 15/22 (2006.01); G06F 40/284 (2020.01); G06N 3/08 (2023.01); G10L 15/06 (2013.01); G10L 15/18 (2013.01); G10L 15/20 (2006.01)
CPC G10L 15/22 (2013.01) [G06F 40/284 (2020.01); G06N 3/08 (2013.01); G10L 15/063 (2013.01); G10L 15/1815 (2013.01); G10L 15/20 (2013.01); G10L 2015/223 (2013.01)] 16 Claims
OG exemplary drawing
 
1. A method implemented by one or more processors, the method comprising:
receiving a user interface input from a user during a human-to-computer dialog session between the user and an automated assistant accessible via a client device;
determining whether to provide, responsive to the user interface input, only a direct response that is responsive to the user interface input or to instead provide an expanded response that includes the direct response and that further includes a prompt that seeks further engagement of the user in the human-to-computer dialog session, wherein determining whether to provide only the direct response or to instead provide the expanded response comprises:
processing, using a trained machine learning model, dialog inputs of the human-to-computer dialog session and/or historical inputs of past interactions of the user with the client device, to generate an output that indicates desirability of providing expanded responses;
in response to determining, based on the output generated using the trained machine learning model, to provide the expanded response:
generating, by the automated assistant, the expanded response that includes the prompt that seeks further engagement, wherein generating the expanded response comprises generating the expanded response based on applying multiple items of responsive content to a trained generative model,
wherein the multiple items of responsive content are applied based on each being responsive to the user interface input, and
wherein the multiple items of responsive content are from different resources, and
causing the client device to render, responsive to the user interface input, the expanded response that includes the direct response and that includes the prompt that seeks further engagement; and
in response to determining, based on the output generated using the trained machine learning model, to provide only the direct response:
causing the client device to render, responsive to the user interface input, the direct response without rendering the expanded response.