公开(公告)号：US12038918B1

公开(公告)日：2024-07-16

申请号：US18225086

申请日：2023-07-21

Applicant: Intuit Inc.

Inventor： Jiaxin Zhang ,  Kamalika Das ,  Sricharan Kumar

IPC: G06F16/242 ,  G06N3/084

CPC classification number: G06F16/243 ,  G06N3/084

Abstract: Disambiguity in large language models (LLMs) includes receiving an original query in a user interface, generating an ambiguity query from the original query, and sending, via an application programming interface (API) of an LLM, the ambiguity query to the LLM. The ambiguity query includes the original query and training the LLM to recognize ambiguities. The method further includes receiving, via the API and responsive to the ambiguity query, a binary response and detecting, based at least in part on the binary response, the original query as ambiguous. Disambiguity may include detecting an ambiguity location in the original query using perturbed queries and the LLM.

公开(公告)号：US11977978B2

公开(公告)日：2024-05-07

申请号：US16944019

申请日：2020-07-30

Applicant: INTUIT INC.

Inventor： Sambarta Dasgupta ,  Sricharan Kumar ,  Ji Chen ,  Debasish Das

IPC: G06N3/08 ,  G06N3/04

CPC classification number: G06N3/08 ,  G06N3/04

Abstract: Certain aspects of the present disclosure provide techniques for performing finite rank deep kernel learning. In one example, a method for performing finite rank deep kernel learning includes receiving a training dataset; forming a set of embeddings by subjecting the training dataset to a deep neural network; forming, from the set of embeddings, a plurality of dot kernels; linearly combining the plurality of dot kernels to form a composite kernel for a Gaussian process; receiving live data from an application; and predicting a plurality of values and a plurality of uncertainties associated with the plurality of values simultaneously using the composite kernel.

公开(公告)号：US11379726B2

公开(公告)日：2022-07-05

申请号：US16212601

申请日：2018-12-06

Applicant: INTUIT INC.

Inventor： Sambarta Dasgupta ,  Sricharan Kumar ,  Ashok Srivastava

IPC: G06N3/08 ,  G06N3/04 ,  G06K9/62 ,  G06N20/10

Abstract: Certain aspects of the present disclosure provide techniques for performing finite rank deep kernel learning. In one example, a method for performing finite rank deep kernel learning includes receiving a training dataset; forming a set of embeddings by subjecting the training data set to a deep neural network; forming, from the set of embeddings, a plurality of dot kernels; combining the plurality of dot kernels to form a composite kernel for a Gaussian process; receiving live data from an application; and predicting a plurality of values and a plurality of uncertainties associated with the plurality of values simultaneously using the composite kernel.