Google Inc.

Matthew K Gray

Namrata U Godbole

Russell J Ryan

Andrew Stephen Tomkins

Ibrahim Badr

Srinivasan Venkatachary

Jayant Madhavan

John Blitzer

R Michael Angelo

Roni Fischman Zeiger

Gregory Harris Plesur

Grant Dasher

Mahsan Rofouei

Thomas A Lasko

Corinn Herrick

Sam Oates

Aditya Shah

Phiroze Parakh

David Robert Bailey

Nundu Janakiram

Thomas W Colthurst

Garrett H Rooney

Marcin K Wichary

Alison Cichowlas

Steven D Baker

Nathan Dickerson Beach

Eric Steinlauf

Greg Harris Plesur

Samer Hassan Hassan

Marius G Andrei

Ionut Georgescu

Yuan Kui Shen

At least selectively utilizing a large language model (LLM) in generating a natural language (NL) based summary to be rendered in response to a query. In some implementations, in generating the NL based summary additional content is processed using the LLM. The additional content is in addition to query content of the query itself and, in generating the NL based summary, can be processed using the LLM and along with the query content—or even independent of the query content. Processing the additional content can, for example, mitigate occurrences of the NL based summary including inaccuracies and/or can mitigate occurrences of the NL based summary being over-specified and/or under-specified.

Generative summaries for search results

At least selectively utilizing a large language model (LLM) in generating a natural language (NL) based summary to be rendered in response to a query. In some implementations, in generating the NL based summary additional content is processed using the LLM. The additional content is in addition to query content of the query itself and, in generating the NL based summary, can be processed using the LLM and along with the query content&#x2014;or even independent of the query content. Processing the additional content can, for example, mitigate occurrences of the NL based summary including inaccuracies and/or can mitigate occurrences of the NL based summary being over-specified and/or under-specified.

Methods and systems for providing an image with an answer to a question query. A method receiving a first query that is determined to be a question query; generating a second query form the first query; causing a search of a plurality of resources of a first corpus using the second query as input, wherein the first corpus of resources are resources that include images and textual data rendered with the images; receiving data indicating a set of resources determined to be responsive to the second query; causing a search for images in the set of resources; selecting, from the set of images and based in part on the search for images in the first set of resources, an image to be provided with the answer in response to the query; and providing, by the data processing apparatus, and with the answer, the image in response to the first query.

Images for query answers

Methods, systems, and apparatus, including computer programs encoded on computer storage media, for triggering rich results in response to queries. In one aspect, a method includes receiving a query. One or more search results are obtained from a first corpus. A rich result is triggered based on a score of the first-ranked search result if it meets a threshold relative to other search results. The rich result is populated with additional metadata about the first-ranked search result obtained from a second corpus. The rich result is provided in response to the query.

Rich results relevant to user search queries

The technology described relates to animated transitions between consecutive sets of search engine results.

Method and apparatus for animating transitions between search results

Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for identifying an image for an entity. In one aspect, a method includes identifying a set of resources. Each resource can include at least one image and reference at least one entity. Each image is assigned to a particular entity based on an overall entity scores for the image relative to the entities. The overall entity score for an image and an entity can specify a measure of topical relatedness between the image and the entity. For each individual entity referenced by at least one of the resources, a group of images that have been assigned to the individual entity is identified. An image evocative of each individual entity is selected from the group based on image rank scores. The image rank score for an image can be determined based on visual characteristics of the image.

Identifying an image for an entity

Methods, systems, and apparatus, including computer programs encoded on computer storage media, for triggering rich results in response to publication queries. In one aspect, a method includes receiving a publication query. One or more publication search results are obtained. A rich result is triggered based on a score of the first-ranked publication result if it meets a threshold relative to other publication search results. The rich result is populated with additional metadata about the publication obtained from web resources and product resources from multiple corpora. The rich result is provided with the publication search results.

Rich results relevant to user search queries for books

Characteristics of a book are determined. A plurality of maximum-likelihood dates associated with text pages within the book is accessed. A maximum-likelihood date for a text page indicates a time period during which that text page was most likely to have been written. The plurality of maximum-likelihood dates is ordered according to the maximum-likelihood dates' associated text pages, from a first text page for which a maximum-likelihood date is available to a last text page for which a maximum-likelihood date is available. The ordered plurality of maximum-likelihood dates is analyzed according to a heuristic to determine characteristics of the book.

Determining book characteristics based on text sections

Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for providing image search results with web corpus search results. In one aspect, a method includes determining dimensions of an image display environment in which image search results are to be displayed, determining an insertion score that defines an ordinal insertion position at which the image display environment is to be inserted into a ranking of general search results that defines ordinal positions for each of the general search results, the dimensions and the insertion score determined, in part, from the quality scores of the images, generating a search results resource for displaying the general search results according to their respective ordinal positions and the image display environment at the ordinal insertion position on a client device, and providing the search results resource in response to the search query.

Dynamic image display area and image display within web search results

Methods, systems, and apparatus, including computer programs encoded on computer storage media, for identifying entities using search results. One of the methods includes determining that a first search query includes a respective text reference to each of one or more predetermined attributes, wherein each attribute is associated with a first entity type; obtaining search results for the first search query from a search engine, the search results identifying a plurality of resources; identifying entities of the first entity type that are related to any of the plurality of resources identified by the search results; and selecting names of one or more of the identified entities of the first entity type to include in a response to the first search query.

Identifying entities using search results

Methods, systems, and apparatus, including computer programs encoded on computer storage media, for identifying entities using search results. One of the methods includes determining that a search query includes a respective text reference to each of one or more predetermined attributes, wherein each attribute is associated with an entity type; obtaining search results for the search query from a search engine, each search result identifying a respective resource; for each of a plurality of the obtained search results, determining an initial score for each of a plurality of entities of the entity type based on occurrences of names of the entity in the resource identified by the search result; generating a final score for each of the entities based on the initial scores; and selecting one or more names of entities of the entity type to include in a response to the search query based on the final scores.

Growing community of inventors

Reading, MA, United States of America

Matthew K Gray