Machine-learned language models which generate intermediate textual analysis in service of contextual text generation

Patent metadata

Field	Value
Publication number	US-11960848-B2
Application number	US-202318164216-A
Country	US
Kind code	B2
Filing date	Feb 3, 2023
Priority date	May 21, 2021
Publication date	Apr 16, 2024
Grant date	Apr 16, 2024

Abstract

The present disclosure is directed to systems and methods that include and/or leverage one or more machine-learned language models that generate intermediate textual analysis (e.g., including usage of structural tools such as APIs) in service of contextual text generation. For example, a computing system can obtain a contextual text string that includes one or more contextual text tokens. The computing system can process the contextual text string with the machine-learned language model to generate one or more intermediate text strings that include one or more intermediate text tokens. The computing system can process the one or more intermediate text strings with the machine-learned language model to generate an output text string comprising one or more output text tokens. The one or more intermediate text strings can include textual analysis of the contextual text string that supports the output text string.

First claim

What is claimed is:

1. A computing system for operating a machine-learned language model to interact with a search engine to generate output text based on input text, the computing system comprising:
one or more processors;
one or more non-transitory computer-readable media that collectively store:
a machine-learned language model; and
instructions that, when executed by the one or more processors, cause the computing system to perform operations, the operations comprising:
obtaining, by the computing system comprising one or more computing devices, the input text;
processing, by the computing system, the input text with the machine-learned language model to generate intermediate text as an intermediate output of the machine-learned language model, wherein the intermediate text invokes a query service to query results from the search engine;
obtaining, by the computing system, the results from the search engine;
processing, by the computing system, at least one of the results from the search engine with the machine-learned language model to generate the output text; and
providing, by the computing system, the output text as an output.

2. The computing system of claim 1 , wherein the input text comprises a question input by a user, and wherein the output text comprises an answer to the question.

3. The computing system of claim 1 , wherein:
the operations further comprise appending the at least one of the results from the search engine to the input text to generate updated text; and
processing, by the computing system, the at least one of the results from the search engine with the machine-learned language model to generate the output text comprises processing, by the computing system, the updated text with the machine-learned language model to generate the output text.

4. The computing system of claim 1 , wherein at least a portion of the intermediate text is structured according to an application programming interface associated with the search engine.

5. The computing system of claim 1 , wherein the machine-learned language model comprises a transformer model.

6. The computing system of claim 1 , wherein the input text comprises text input by a user.

7. The computing system of claim 1 , wherein the machine-learned language model has been trained using reinforcement learning.

8. A computer-implemented method for operating a machine-learned language model to interact with a search engine to generate output text based on input text, the method comprising:
obtaining, by a computing system comprising one or more computing devices, the input text;
processing, by the computing system, the input text with the machine-learned language model to generate intermediate text as an intermediate output of the machine-learned language model, wherein the intermediate text invokes a query service to query results from the search engine;
obtaining, by the computing system, the results from the search engine;
processing, by the computing system, at least one of the results from the search engine with the machine-learned language model to generate the output text; and
providing, by the computing system, the output text as an output.

9. The computer-implemented method of claim 8 , wherein the input text comprises a question input by a user, and wherein the output text comprises an answer to the question.

10. The computer-implemented method of claim 8 , wherein:
the method further comprises appending the at least one of the results from the search engine to the input text to generate updated text; and
processing, by the computing system, the at least one of the results from the search engine with the machine-learned language model to generate the output text comprises processing, by the computing system, the updated text with the machine-learned language model to generate the output text.

11. The computer-implemented method of claim 8 , wherein at least a portion of the intermediate text is structured according to an application programming interface associated with the search engine.

12. The computer-implemented method of claim 8 , wherein the machine-learned language model comprises a transformer model.

13. The computer-implemented method of claim 8 , wherein the input text comprises text input by a user.

14. The computer-implemented method of claim 8 , wherein the machine-learned language model has been trained using reinforcement learning.

15. One or more non-transitory computer-readable media that collectively store instructions that, when executed by one or more processors of a computing system, cause the computing system to perform operations, the operations comprising:
obtaining, by the computing system, an input text;
processing, by the computing system, the input text with a machine-learned language model to generate intermediate text as an intermediate output of the machine-learned language model, wherein the intermediate text invokes a query service to query results from the search engine;
obtaining, by the computing system, the results from the search engine;
processing, by the computing system, at least one of the results from the search engine with the machine-learned language model to generate an output text; and
providing, by the computing system, the output text as an output.

16. The one or more non-transitory computer-readable media of claim 15 , wherein the input text comprises a question input by a user, and wherein the output text comprises an answer to the question.

17. The one or more non-transitory computer-readable media of claim 15 , wherein:
the operations further comprise appending the at least one of the results from the search engine to the input text to generate updated text; and
processing, by the computing system, the at least one of the results from the search engine with the machine-learned language model to generate the output text comprises processing, by the computing system, the updated text with the machine-learned language model to generate the output text.

18. The one or more non-transitory computer-readable media of claim 15 , wherein at least a portion of the intermediate text is structured according to an application programming interface associated with the search engine.

19. The one or more non-transitory computer-readable media of claim 15 , wherein the input text comprises text input by a user.

20. The one or more non-transitory computer-readable media of claim 15 , wherein the machine-learned language model has been trained using reinforcement learning.

Assignees

• Google Llc

Inventors

• Shazeer Noam
• De Freitas Adiwardana Daniel

Classifications

• G06F40/35 Primary

  Discourse or dialogue representation · CPC title

• G06N3/045

  Combinations of networks · CPC title

• G06F16/90332

  Natural language query formulation or dialogue systems · CPC title

• G06F40/284 Primary

  Lexical analysis, e.g. tokenisation or collocates · CPC title

• G06F40/279

  Recognition of textual entities · CPC title