Generative summaries for search results

1. A system, comprising:

memory storing instructions;

one or more processors operable to execute the instruction to:

receive a query associated with a client device;

in response to receiving the query:

select a set of search result documents, wherein in selecting the set of search result documents one or more of the processors are to select, for inclusion in the set, a plurality of query-responsive search result documents based on the query-responsive search result documents being responsive to the query;

generate large language model (LLM) output based on processing, using an LLM, the query and corresponding content from each of the search result documents of the set of search result documents;

generate a natural language (NL) based summary using the LLM output;

determine that a first portion of content, of the NL based summary, is verifiable based on first document content of a first document of the set of search result documents;

determine that a second portion of content, of the NL based summary, is verifiable based on second document content of a second document of the set of search result documents, the second portion of content being disparate from the first portion of content;

determine that the first portion of content, of the NL based summary, is also verifiable based on third document content of a third document;

cause the NL based summary to be rendered at the client device, wherein in causing the NL based summary to be rendered at the client device one or more of the processors are to:

cause a first selectable graphical element to be rendered immediately after the first portion of content, wherein the first selectable graphical element, when selected, causes navigation to the first document, and wherein the first selectable graphical element is rendered immediately after the first portion of content responsive to determining that the first portion of content is verifiable based on the first document content of the first document,

cause a second selectable graphical element to be rendered immediately after the second portion of content, wherein the second selectable graphical element, when selected, causes navigation to the second document, and wherein the second selectable graphical element is rendered immediately after the second portion of the content responsive to determining that the second portion of the content is verifiable based on the second document content of the second document, and

cause a third selectable graphical element to be rendered immediately after the first portion of content, wherein the third selectable graphical element, when selected, causes navigation to the third document, and wherein the third selectable graphical element is rendered immediately after the first portion of content responsive to determining that the first portion of content is also verifiable based on the third document content of the third document.

2. The system of claim 1 , wherein the first search result document and the second search result document are among the query-responsive search result documents that are responsive to the query.

3. The system of claim 1 , wherein in determining that the first portion of content is verifiable based on first document content of the first document one or more of the processors are to:

compare the first portion of content to the first document content; and

determine, based on the comparing, that the first portion of content is verifiable based on the first document content of the first document.

4. The system of claim 3 ,

wherein in comparing the first portion of content to the first document content one or more of the processors are to:

process the first portion of content, using an encoder model, to generate a content embedding of the first portion of content; and

process the first document content, using the encoder model, to generate a document content embedding of the first document content; and

wherein in determining, based on the comparing, that the first portion of content is verifiable based on the first document content of the first document one or more of the processors are to:

determine a distance measure between the content embedding and the document content embedding; and

determine, based on the distance measure, that the first portion of content is verifiable based on the first document content of the first document.

5. The system of claim 1 , wherein one or more of the processors are further operable to execute the instructions to:

in response to receiving the query:

cause additional search results to be rendered at the client device along with, but separate from, the NL based summary,

wherein in causing the additional search results to be rendered, one or more of the processors are to cause the additional search results to be rendered based on the additional search results being responsive to the query, and

wherein the additional search results are for corresponding search result documents that are in addition to those of the set of search result documents.

6. The system of claim 1 , wherein the NL based summary is initially rendered without also rendering additional search results that are for corresponding search result documents that are responsive to the query and that are in addition to those of the set of search result documents.

7. The system of claim 6 , wherein one or more of the processors are further operable to execute the instructions to:

cause a selectable additional search result element to be rendered along with initial rendering of the NL based summary; and

in response to a selection of the selectable additional search result element:

cause the additional search results to be rendered at the client device along with, but separate from, the NL based summary.

8. The system of claim 1 , wherein one or more of the processors are further operable to execute the instructions to:

prior to generating the LLM output:

determine whether to utilize the LLM in generating one or more responses to render responsive to receiving the query;

wherein generating the LLM output, using the LLM, is responsive to determining to utilize the LLM in generating one or more responses to render responsive to receiving the query.

9. The system of claim 8 , wherein in determining whether to utilize the LLM in generating one or more responses to render responsive to receiving the query one or more of the processors are to:

determine, based on one or more query features of the query and/or based on one or more result features, whether to utilize the LLM, wherein the one or more result features are based on one or more of the query-responsive search result documents that are responsive to the query.

10. The system of claim 1 , wherein in generating the LLM output based on processing, using the LLM, the query and the corresponding content from each of the search result documents of the set of search result documents one or more of the processors are to:

generate a prompt that includes the query, the corresponding content, and one or more additional terms that are in addition to those of the query and the corresponding content; and

process the prompt, using the LLM, to generate the LLM output.

11. The system of claim 1 , wherein the query associated with the client device is automatically generated based on one or more of a profile of a user of the client device, a history of the user, and a context of the user.

12. A method implemented by one or more processors, the method comprising:

receiving a query formulated based on user interface input at a client device;

in response to receiving the query:

selecting a set of search result documents, selecting the set of search result documents including selecting, for inclusion in the set, a plurality of query-responsive search result documents based on the query-responsive search result documents being responsive to the query;

generating a prompt based on the query and based on corresponding content from each of the search result documents of the set of search result documents;

causing generation of a natural language (NL) based summary, causing generation of the NL based summary including causing the prompt to be processed, utilizing a large language model (LLM), to generate LLM output that reflects the NL based summary;

determining that a first portion of content, of the NL based summary, is verifiable based on first document content of a first document of the set of search result documents;

determining that a second portion of content, of the NL based summary, is verifiable based on second document content of a second document of the set of search result documents, the second portion of content being disparate from the first portion of content;

determining that the first portion of content, of the NL based summary, is also verifiable based on third document content of a third document;

causing the NL based summary to be rendered at the client device, comprising:

causing a first selectable graphical element to be rendered immediately after the first portion of content, wherein the first selectable graphical element, when selected, causes navigation to the first document, and wherein the first selectable graphical element is rendered immediately after the first portion of content responsive to determining that the first portion of content is verifiable based on the first document content of the first document,

causing a second selectable graphical element to be rendered immediately after the second portion of content, wherein the second selectable graphical element, when selected, causes navigation to the second document, and wherein the second selectable graphical element is rendered immediately after the second portion of the content responsive to determining that the second portion of the content is verifiable based on the second document content of the second document, and

causing a third selectable graphical element to be rendered immediately after the first portion of content, wherein the third selectable graphical element, when selected, causes navigation to the third document, and wherein the third selectable graphical element is rendered immediately after the first portion of content responsive to determining that the first portion of content is also verifiable based on the third document content of the third document.

13. The method of claim 12 , wherein the third document is included in the set of search result documents.

14. The method of claim 12 , wherein the first search result document and the second search result document are among the query-responsive search result documents that are responsive to the query.

15. The method of claim 12 , wherein determining that the first portion of content is verifiable based on first document content of the first document comprises:

comparing the first portion of content to the first document content; and

determining, based on the comparing, that the first portion of content is verifiable based on the first document content of the first document.

16. The method of claim 15 ,

wherein comparing the first portion of content to the first document content comprises:

processing the first portion of content, using an encoder model, to generate a content embedding of the first portion of content; and

processing the first document content, using the encoder model, to generate a document content embedding of the first document content; and

wherein determining, based on the comparing, that the first portion of content is verifiable based on the first document content of the first document comprises:

determining a distance measure between the content embedding and the document content embedding; and

determining, based on the distance measure, that the first portion of content is verifiable based on the first document content of the first document.

17. The method of claim 12 , further comprising:

in response to receiving the query:

causing additional search results to be rendered at the client device along with, but separate from, the NL based summary,

wherein causing the additional search results to be rendered is based on the additional search results being responsive to the query, and

wherein the additional search results are for corresponding search result documents that are in addition to those of the set of search result documents.

18. The method of claim 12 , wherein the NL based summary is initially rendered without also rendering additional search results that are for corresponding search result documents that are responsive to the query and that are in addition to those of the set of search result documents.

19. The method of claim 12 , further comprising:

prior to causing generation of the NL based summary:

determining whether to utilize the LLM in generating one or more responses to render responsive to receiving the query;

wherein causing generation of the NL based summary is responsive to determining to utilize the LLM in generating one or more responses to render responsive to receiving the query.

20. A system, comprising:

memory storing instructions;

one or more processors operable to execute the instruction to:

receive a query associated with a client device;

in response to receiving the query:

select a set of search result documents, wherein in selecting the set of search result documents one or more of the processors are to select, for inclusion in the set, a plurality of query-responsive search result documents based on the query-responsive search result documents being responsive to the query;

generate large language model (LLM) output based on processing, using an LLM, the query and corresponding content from each of the search result documents of the set of search result documents, wherein in generating the LLM output based on processing, using the LLM, the query and the corresponding content from each of the search result documents of the set of search result documents one or more of the processors are to:

generate a prompt that includes the query, the corresponding content, and one or more additional terms that are in addition to those of the query and the corresponding content; and

process the prompt, using the LLM, to generate the LLM output;

generate a natural language (NL) based summary using the LLM output;

determine that a first portion of content, of the NL based summary, is verifiable based on first document content of a first document of the set of search result documents;

determine that a second portion of content, of the NL based summary, is verifiable based on second document content of a second document of the set of search result documents, the second portion of content being disparate from the first portion of content;

cause the NL based summary to be rendered at the client device, wherein in causing the NL based summary to be rendered at the client device one or more of the processors are to:

cause a first selectable graphical element to be rendered immediately after the first portion of content, wherein the first selectable graphical element, when selected, causes navigation to the first document, and wherein the first selectable graphical element is rendered immediately after the first portion of content responsive to determining that the first portion of content is verifiable based on the first document content of the first document, and

cause a second selectable graphical element to be rendered immediately after the second portion of content, wherein the second selectable graphical element, when selected, causes navigation to the second document, and wherein the second selectable graphical element is rendered immediately after the second portion of the content responsive to determining that the second portion of the content is verifiable based on the second document content of the second document.

21. The system of claim 20 , wherein the first search result document and the second search result document are among the query-responsive search result documents that are responsive to the query.

22. The system of claim 20 , wherein one or more of the processors are further operable to execute the instructions to:

in response to receiving the query:

cause additional search results to be rendered at the client device along with, but separate from, the NL based summary,

wherein in causing the additional search results to be rendered, one or more of the processors are to cause the additional search results to be rendered based on the additional search results being responsive to the query, and

wherein the additional search results are for corresponding search result documents that are in addition to those of the set of search result documents.

23. The system of claim 20 , wherein the NL based summary is initially rendered without also rendering additional search results that are for corresponding search result documents that are responsive to the query and that are in addition to those of the set of search result documents.

24. The system of claim 20 , wherein one or more of the processors are further operable to execute the instructions to:

prior to generating the LLM output:

determine whether to utilize the LLM in generating one or more responses to render responsive to receiving the query;

wherein generating the LLM output, using the LLM, is responsive to determining to utilize the LLM in generating one or more responses to render responsive to receiving the query.