公开(公告)号：US11861315B2

公开(公告)日：2024-01-02

申请号：US17351501

申请日：2021-06-18

Applicant: Meta Platforms, Inc.

Inventor： Pooja Sethi ,  Denis Savenkov ,  Yue Liu ,  Alexander Kolmykov-Zotov ,  Ahmed Aly

IPC: G06F17/00 ,  G06F40/30 ,  G06F1/3206 ,  G06F3/01 ,  G06F3/04815 ,  G06N5/02 ,  G06N5/046 ,  G06T19/00 ,  G06T19/20

CPC classification number: G06F40/30 ,  G06F1/3206 ,  G06F3/011 ,  G06F3/04815 ,  G06N5/02 ,  G06N5/046 ,  G06T19/006 ,  G06T19/20 ,  G06T2219/2004

Abstract: In one embodiment, a method includes receiving a user request to automatically debug a natural-language understanding (NLU) model, accessing a plurality of predicted semantic representations generated by the NLU model, wherein the plurality of predicted semantic representations are associated with a plurality of dialog sessions, respectively, wherein each dialog session is between a user and an assistant xbot associated with the NLU model, generating a plurality of expected semantic representations associated with the plurality of dialog sessions based on an auto-correction model, wherein the auto-correction model is learned from dialog training samples generated based on active learning, identifying incorrect semantic representations of the predicted semantic representations based on a comparison between the predicted semantic representations and the expected semantic representations, and automatically correcting the incorrect semantic representations by replacing them with respective expected semantic representations generated by the auto-correction model.

公开(公告)号：US20220374605A1

公开(公告)日：2022-11-24

申请号：US17351501

申请日：2021-06-18

Applicant: Meta Platforms, Inc.

Inventor： Pooja Sethi ,  Denis Savenkov ,  Yue Liu ,  Alexander Kolmykov-Zotov ,  Ahmed Aly

IPC: G06F40/30

Abstract: In one embodiment, a method includes receiving a user request to automatically debug a natural-language understanding (NLU) model, accessing a plurality of predicted semantic representations generated by the NLU model, wherein the plurality of predicted semantic representations are associated with a plurality of dialog sessions, respectively, wherein each dialog session is between a user and an assistant xbot associated with the NLU model, generating a plurality of expected semantic representations associated with the plurality of dialog sessions based on an auto-correction model, wherein the auto-correction model is learned from dialog training samples generated based on active learning, identifying incorrect semantic representations of the predicted semantic representations based on a comparison between the predicted semantic representations and the expected semantic representations, and automatically correcting the incorrect semantic representations by replacing them with respective expected semantic representations generated by the auto-correction model.