Microsoft Technology Licensing, LLC

Raghavan Muthuregunathan

Abhimanyu Lad

Santhosh Sachindran

Manish Malik

Doran Chakraborty

Sanjay Reddy Satti

Yuankun Xue

Xu Zhang

Jiayu Li

Poome Thavornvanit

Yi Zhou

Vivek Katarya

Jagadeesan Sundaresan

Dhruv Saksena

Ali Hooshmand

Chetan Bhole

Techniques are provided for presenting a summary with search results. One method includes operations for receiving a search query, performing a search based on the search query, and determining, by a machine learning (ML) model, a user intent based on input comprising the search query, user profile information, and activity information of the user. A prompt is generated based on the search query, the user intent, and relationships between the user and entities. The prompt generation comprises selecting a prompt template comprising fields for the user intent, the search query, and the relationships, and filling in the fields. The method further includes operations for feeding as input the prompt to a generative artificial intelligence (GAI) model, receiving a summary from the GAI model, and causing presentation of the summary and one or more results returned by the search on a user interface (UI) of the device of the user.

Search summary generation based on searcher characteristics

Implementations of the disclosed technologies pre-fetch search results. Implementations receive first input from a search session of a user device, where the first input includes at least a portion of a search term but does not initiate a search. Implementations determine context data associated with the first input, determine that a combination of the first input and the context data satisfies a pre-fetch threshold, determine intent data based on at least a portion of the context data, generate a search query based on the first input and the intent data, and pre-fetch a first subset of search results based on the search query. In response to a second input received subsequent to the first input, where the second input contains an initiate search signal, implementations initiate rendering of the pre-fetched first subset of search results in the search session at the user device.

Instant search results

In various example embodiments, a system and method for modifying user input are presented. In one embodiment, a method includes receiving electronic input from a member of a social networking service, the input comprising a named entity; transmitting a query to a database of entities connected to the social networking service to select a set of entities in the database whose names are within a character edit distance of a spelling of the named entity; identifying a matching entity that is connected to the member at the social networking service and matches an entity in the set of entities; and modifying the input by replacing the spelling of the named entity with a spelling of the matching entity.

Querying named entities

Disclosed are systems, methods, and non-transitory computer-readable media for expanding search queries using query term weighting. A search system receives a search term comprising multiple keywords and executes a search query based on the search term. In response to determining that the number of search results is less than a threshold number, the search system determines, based on historical query reformulation data, weight values for the keywords included in the search term. The weight values indicate a probability that each respective keyword would be removed when reformulating the search term. The search system generates a reformulated search term based on the weight values and executes a second search query based on the reformulated search term. The search system causes presentation of at least a portion of the search results on a client device of the requesting user.

Expanding search queries using query term weighting

Aspects of the technology described herein increase the efficiency of a search session by determining whether fresh content is likely to be responsive to the user's query. Whether fresh content is likely to be responsive to a specific query is determined by retrieving social media posts that are responsive to the query. The social media posts are evaluated for virality, which is the tendency of a social media post to be circulated rapidly and widely from one Internet user to another. The virality of a social media post can be determined by comparing a number of times the social media post has been re-communicated by individual users. Queries that return viral social media posts may be classified as seeking fresh content.

Efficient retrieval of fresh internet content

Growing community of inventors

Sunnyvale, CA, United States of America

Raghavan Muthuregunathan