Salesforce, Inc. patent applications on April 17th, 2025
Patent Applications by Salesforce, Inc. on April 17th, 2025
Salesforce, Inc.: 5 patent applications
Salesforce, Inc. has applied for patents in the areas of G06F9/50 (1), G06F11/20 (1), G06F11/14 (1), G06F16/27 (1), G06F16/242 (1) G06F9/5038 (1), G06F11/2023 (1), G06F16/243 (1), G06F40/20 (1), G06F40/40 (1)
With keywords such as: llm, data, include, server, language, large, intermediate, based, node, and information in patent application abstracts.
Patent Applications by Salesforce, Inc.
Inventor(s): Dustin Allen Kirk of San Diego CA US for salesforce, inc.
IPC Code(s): G06F9/50
CPC Code(s): G06F9/5038
Abstract: a method of data processing is described. the method may include receiving an indication of one or more operational constraints for an unstructured compute process to be executed by the data processing system. the method may further include transmitting a request to decompose the unstructured compute process into a set of structured sub-processes and allocate the set of structured sub-processes to a set of compute resources of the data processing system in accordance with the one or more operational constraints. the method may further include receiving a response that contains information associated with an allocation scheme for the set of structured sub-processes of the unstructured compute process. the method may further include allocating the set of structured sub-processes to the set of compute resources and executing the set of structured sub-processes in accordance with the allocation scheme and the one or more operational constraints.
Inventor(s): Damon Ashman of Indianapolis IN US for salesforce, inc., Vi Nguyen of Bellevue WA US for salesforce, inc., Rishi Zaveri of West Bloomfield MI US for salesforce, inc., Arockia Praveen Thargis of Indianapolis IN US for salesforce, inc., Omar Jaber of Bellevue WA US for salesforce, inc., Brennan Lindamood of Attica IN US for salesforce, inc., Bradley Kenneth Michel of San Francisco CA US for salesforce, inc.
IPC Code(s): G06F11/20, G06F11/14, G06F16/27
CPC Code(s): G06F11/2023
Abstract: systems and methods are provided for replacing a first server of a first node a computer network by adding a second server configured with a database system as a second node to the computer network. a snapshot of a first database of database files that is communicatively coupled to the first node may be taken. the snapshot may be applied to the second server by attaching the snapshot of the first database as a second database to the second server and configuring the second server to match the first server. a failover of the computer network to the second node with the second server having the second database attached may be forced. voting may be switched to the second server of the second node of the computer network and the first node of the computer network may be removed when the forced failover is successful.
20250124024. GENERATED CONTENT SOURCE_simplified_abstract_(salesforce, inc.)
Inventor(s): Dustin Allen Kirk of San Diego CA US for salesforce, inc.
IPC Code(s): G06F16/242, G06F16/248, G06F21/60
CPC Code(s): G06F16/243
Abstract: a large language model (llm) may receive a query via a user interface of a client device. the llm may generate one or more data queries from the query to query one or more data sets. the llm may then transmit the one or more data queries to the one or more data sets. the llm may then receive information associated with the query and a source for the information from the one or more data sets. the source may be indicative of a location within the one or more data sets from where the information was obtained. following, the llm may generate a response to the query that includes the information associated with the query and the source for the information and transmit the response to the user interface of the client device for display.
20250124223. MULTI-LAYERED CUSTOMIZATION FRAMEWORK_simplified_abstract_(salesforce, inc.)
Inventor(s): Dustin Allen Kirk of San Diego CA US for salesforce, inc.
IPC Code(s): G06F40/20
CPC Code(s): G06F40/20
Abstract: methods, apparatuses, and computer-program products are disclosed. the method may include deriving a first intermediate llm based on a broad spectrum llm. the first intermediate llm may be associated with a first level of a hierarchy and deriving the first intermediate llm may include training the first intermediate llm on first training data associated with the first level of the hierarchy. the method may include deriving a user-level llm based on the first intermediate llm, where the user-level llm is associated with a user associated with a second level of the hierarchy included in the first level of the hierarchy. deriving the user-level llm may include training the user-level llm on second training data associated with the user, where deriving the user-level llm may further include inheriting one or more first characteristics from the first intermediate llm.
Inventor(s): Itai Izhak Feigenbaum of San Francisco CA US for salesforce, inc., Devansh Arpit of Pacifica CA US for salesforce, inc., Shelby Heinecke of San Francisco CA US for salesforce, inc., Juan Carlos Niebles Duque of Mountain View CA US for salesforce, inc., Weiran Yao of San Francisco CA US for salesforce, inc., Huan Wang of Palo Alto CA US for salesforce, inc., Caiming Xiong of Menlo Park CA US for salesforce, inc., Silvio Savarese of Palo Alto CA US for salesforce, inc.
IPC Code(s): G06F40/40, G06N7/01
CPC Code(s): G06F40/40
Abstract: systems and methods for editing a large language model are provided. the large language model generates a sequence of tokens, a first probability of a pre-edit output based on the sequence of tokens, and a second probability of a target output based on the sequence of tokens. a loss function is provided based on the first probability and the second probability. a plurality of gradients of the large language model with respect to the loss function is computed. an edit location of the large language model is determined based on the plurality of gradients. the large language model is edited by editing weights at the edit location of the large language model, such that the updated large language model generates the target output for an input including the sequence of words.