Microsoft Technology Licensing, LLC

Wenjin Xu

Xiao Li

Yujie He

Nikita Voronkov

Sebastian Johannes Blohm

Kun Piao

Hadi Kotaich

Dion Stephan Javellana Ong

Qiwen Guo

Jamie Olson

Li Li

Frederick Qin

Rui Min

Aspects of the disclosure include methods and systems for an intelligent chat powered by a large language model that leverages both public and private data to answer user questions. An exemplary method includes receiving a user query including natural language input from a user and executing the user query against at least one public data source and at least one private data source. Queries executed against a public source are retrieved using public search indices and queries executed against a private data source are retrieved using user credentials. A query rewrite and a query context including the user query and retrieved information from the public and private data sources are input to a large language model. A response is received from the large language model that includes a natural language answer to the user query and a link to the retrieved information.

Large language model architecture to leverage public and private data

A data processing system implements receiving a textual context inserted into a user interface element; receiving an indicator inserted into the user interface element after the textual context, the indicator indicating a desire to tag a topic from a plurality of topics included in a knowledge base; receiving one or more textual character inserted into the user interface element after the indicator; encoding, using a machine-learning (ML) model, the received textual context to generate at least one representation reflecting one or more meanings of the received textual context; decoding, using the ML model, the at least one representation to generate a plurality of tokens in response to the one or more meanings of the received textual context, the plurality of tokens corresponding with the at least one textual character and at least one of the topics of the plurality of topics; identifying one or more topics from the plurality of topics as recommended topics; and providing the identified recommended topics for display in a topic selection user interface element that enables selection of one recommended topic for insertion as the tag in the user interface element.

Contextual-based knowledge entity suggestions

A computing system obtains computer-readable text comprising an identifier for an entity and an additional word. The computing system computes relevance scores for entries for entities in a computer-implemented knowledge graph upon the text, titles for the entities in the knowledge graph, and descriptions of the entities in the knowledge graph. The computing system determines whether the identifier for the entity is an acronym or a non-acronym. The computing system identifies an entry for the entity in the knowledge graph based upon the relevance scores and whether the identifier for the entity is an acronym or a non-acronym. The computing system generates a link between the identifier for the entity in the text and the entry for the entity in the knowledge graph, where data from the entry for the entity in the knowledge graph is presented to a user upon the link being selected.

Computing system for entity disambiguation and not-in-list entity detection in a knowledge graph

Growing community of inventors

Bothell, WA, United States of America

Wenjin Xu