公开(公告)号：US11734600B2

公开(公告)日：2023-08-22

申请号：US16376254

申请日：2019-04-05

Applicant: Google LLC

Inventor： Wei Wang ,  Bowen Liang ,  Macduff Hughes ,  Taro Watanabe ,  Tetsuji Nakagawa ,  Alexander Rudnick

IPC: G06N20/00 ,  G06N7/01 ,  H04L67/10

CPC classification number: G06N20/00 ,  G06N7/01 ,  H04L67/10

Abstract: A method includes generating a base model by training with a first dataset of data pairs and generating an adapted model by training the base model on a second dataset of data pairs. The method also includes determining a contrastive score for each data pair of a third dataset of data pairs using the base model and the adapted model. The contrastive score is indicative of a probability of quality of the respective data pair. The method also includes training a target model using the data pairs of the third dataset and the contrastive scores.

公开(公告)号：US20230359938A1

公开(公告)日：2023-11-09

申请号：US18351397

申请日：2023-07-12

Applicant: Google LLC

Inventor： Wei Wang ,  Bowen Liang ,  Macduff Hughes ,  Taro Watanabe ,  Tetsuji Nakagawa ,  Alexander Rudnick

IPC: G06N20/00 ,  G06N7/01

CPC classification number: G06N20/00 ,  G06N7/01 ,  H04L67/10

Abstract: A method includes generating a base model by training with a first dataset of data pairs and generating an adapted model by training the base model on a second dataset of data pairs. The method also includes determining a contrastive score for each data pair of a third dataset of data pairs using the base model and the adapted model. The contrastive score is indicative of a probability of quality of the respective data pair. The method also includes training a target model using the data pairs of the third dataset and the contrastive scores.

公开(公告)号：US20230025739A1

公开(公告)日：2023-01-26

申请号：US17852863

申请日：2022-06-29

Applicant: Google LLC

Inventor： Junpei Zhou ,  Yuezhang Li ,  Ciprian Chelba ,  Fangxiaoyu Feng ,  Bowen Liang ,  Pidong Wang

IPC: G06F40/58 ,  G06N3/08 ,  G06F40/51

Abstract: Aspects of the technology employ a machine translation quality prediction (MTQP) model to refine datasets that are used in training machine translation systems. This includes receiving, by a machine translation quality prediction model, a sentence pair of a source sentence and a translated output (802). Then performing feature extraction on the sentence pair using a set of two or more feature extractors, where each feature extractor generates a corresponding feature vector (804). The corresponding feature vectors from the set of feature extractors are concatenated together (806). And the concatenated feature vectors are applied to a feedforward neural network, in which the feedforward neural network generates a machine translation quality prediction score for the translated output (808).