-
1.发明公开
公开(公告)号：EP4174715A1
公开(公告)日：2023-05-03
申请号：EP22190600.1
申请日：2022-08-16
发明人： SHANG, Junyuan , WANG, Shuohuan , DING, Siyu , ZHAO, Yanbin , PANG, Chao , SUN, Yu
IPC分类号： G06F40/30 , G06F40/284 , G06F40/56 , G06N3/08
摘要： The present disclosure provides a method and apparatus for pre-training a model, a device, a storage medium, and a program product, and relates to the technical field of artificial intelligence, in particular to the technical fields of natural language processing and deep learning. An embodiment of the method includes: acquiring a sample natural language text; generating N types of prompt words based on the sample natural language text, where N is a positive integer; generating sample input data based on the sample natural language text and the N types of prompt words; and training an initial language model based on the sample input data, to obtain a pre-trained language model. This embodiment provides a controllable generation pre-training technology based on prompts, which increases the controllability of the model.
-
公开(公告)号：EP4113387A3
公开(公告)日：2023-03-01
申请号：EP22192187.7
申请日：2022-08-25
发明人： SHI, Hongjian , FENG, Xinwei , LI, Feifei , GUO, Chenyang , WU, Xueqian , TIAN, Meng , SUN, Yu
IPC分类号： G06N3/042 , G06N3/09 , G06N3/045 , G06F16/9038
摘要： The present disclosure provides a search method based on a neural network model, the neural network model including a semantic representation model, a recall model, and a ranking model, and relates to the field of artificial intelligence, and in particular to the technical field of search. An implementation is: inputting a target search and a plurality of objects to be matched to the semantic representation model to obtain a first output of the semantic representation model, where the first output has a semantic understanding representation of recall and ranking; inputting the first output of the semantic representation model to the recall model, and obtaining at least one recall object matching the target search from the plurality of objects to be matched by using the recall model; and inputting a second output of the semantic representation model to the ranking model, and obtaining a matching value of each of the at least one recall object by using the ranking model, where the second output of the semantic representation model is obtained based on the input target search and the at least one recall object.
-
公开(公告)号：EP4113387A2
公开(公告)日：2023-01-04
申请号：EP22192187.7
申请日：2022-08-25
发明人： SHI, Hongjian , FENG, Xinwei , LI, Feifei , GUO, Chenyang , WU, Xueqian , TIAN, Meng , SUN, Yu
IPC分类号： G06N3/04 , G06N3/08 , G06F16/9038
摘要： The present disclosure provides a search method based on a neural network model, the neural network model including a semantic representation model, a recall model, and a ranking model, and relates to the field of artificial intelligence, and in particular to the technical field of search. An implementation is: inputting a target search and a plurality of objects to be matched to the semantic representation model to obtain a first output of the semantic representation model, where the first output has a semantic understanding representation of recall and ranking; inputting the first output of the semantic representation model to the recall model, and obtaining at least one recall object matching the target search from the plurality of objects to be matched by using the recall model; and inputting a second output of the semantic representation model to the ranking model, and obtaining a matching value of each of the at least one recall object by using the ranking model, where the second output of the semantic representation model is obtained based on the input target search and the at least one recall object.
-
公开(公告)号：EP4033415A1
公开(公告)日：2022-07-27
申请号：EP22150836.9
申请日：2022-01-11
发明人： LU, Yuxiang , LIU, Jiaxiang , CHEN, Xuyi , FENG, Shikun , WANG, Shuohuan , SUN, Yu , HUANG, Shiwei , HE, Jingzhou
摘要： A pre-training method of a neural network model, an electronic device, and a medium. The method includes obtaining (601, 701, 801) pre-training data; inputting (602, 702) the pre-training data to an initial neural network model, and pre-training the initial neural network model in a first training mode; obtaining (603, 805) a loss value of the initial neural network model; and when the loss value of the initial neural network model is less than a preset threshold, continuing (604, 706, 806) pre-training the initial neural network model in a second training mode.
-
5.发明公开
公开(公告)号：EP4350577A1
公开(公告)日：2024-04-10
申请号：EP23200782.3
申请日：2023-09-29
IPC分类号： G06N3/045 , G06N3/08 , G06N3/0475
摘要： The present disclosure provides a data generation method based on a deep learning model, and a training method and apparatus, relates to the field of artificial intelligence technologies, and in particular, to the field of natural language processing and deep learning technologies, and can be used to improve the quality of reply data generated by the deep learning model based on input data of a user. The data generation method includes: determining an initial input of the deep learning model based on input data of a user; obtaining a first output of the model, where in response to the model determining that generating a reply based on the initial input requires calling a first functional component different from the deep learning model, the first output includes a first token for calling the first functional component and a first intermediate inquiry determined based on the initial input and recognizable by the first functional component; obtaining a first intermediate result determined by the first functional component based on the first intermediate inquiry; determining a second input for the model based on the initial input and the first intermediate result; and obtaining a second output of the model for generating a reply to the initial input.
-
-
-
-
搜索结果
5 条记录 (耗时 0.013 秒)
