Invention Grant
- Patent Title: Evaluating output sequences using an auto-regressive language model neural network
-
Application No.: US17876451Application Date: 2022-07-28
-
Publication No.: US12086713B2Publication Date: 2024-09-10
- Inventor: Daniel De Freitas Adiwardana , Noam M. Shazeer
- Applicant: Google LLC
- Applicant Address: US CA Mountain View
- Assignee: Google LLC
- Current Assignee: Google LLC
- Current Assignee Address: US CA Mountain View
- Agency: Fish & Richardson P.C.
- Main IPC: G06N3/08
- IPC: G06N3/08 ; G06F18/2113 ; G06N20/00

Abstract:
Methods, systems, and apparatus, including computer programs encoded on computer storage media, for evaluating candidate output sequences using language model neural networks. In particular, an auto-regressive language model neural network is used to generate a candidate output sequence. The same auto-regressive language model neural network is used to evaluate the candidate output sequence to determine rating scores for each of one or more criteria. The rating score(s) are then used to determine whether to provide the candidate output sequence.
Public/Granted literature
- US20230029590A1 EVALUATING OUTPUT SEQUENCES USING AN AUTO-REGRESSIVE LANGUAGE MODEL NEURAL NETWORK Public/Granted day:2023-02-02
Information query