US 2026/0228207 A1 · AUTOMATIC GENERATION OF DATA CARDS

Abstract

In various examples, techniques for automatically generating and maintaining data cards for datasets is described herein. Systems and methods are disclosed that process a dataset in order to identify relevant information associated with the dataset. For example, the dataset may include and/or be associated with sources of information—such as files, documents, links, memos, research papers, annotations, labels, and/or the like—that describe data instances (e.g., images, audio clips, point clouds, etc.) included in the dataset. These sources of information may then be analyzed to retrieve the relevant information associated with the dataset. Systems and methods are then further disclosed that may use one or more language models to process input data associated with the relevant information in order to generate a data card associated with the dataset.

Claims20

Dependent claims

A method comprising:

determining that a dataset has been created for one or more machine learning models to perform one or more tasks;
determining, based at least on analyzing the dataset, first information that is associated with one or more data sources included in the dataset;
generating, based at least on one or more language models processing input data representative of at least the first information, output data representing a data card that includes second information describing the one or more data sources included in the dataset;
and
integrating the data card with one or more platforms associated with the one or more machine learning models.

2 more elements

One or more processors comprising processing circuitry to:

obtain a data card that includes information associated with a dataset;
determine, based at least on the data card, whether the dataset satisfies one or more criteria for performing at least one of training or updating a machine learning model;
and
based at least on determining that the dataset satisfies the one or more criteria, providing the dataset for the at least one of the training or the updating of the machine learning model.

1 more element

Dependencies are inferred from the claim text.

Automatic generation of data cards

Life of this patent

Automatic generation of data cards

Life of this patent

Abstract

Claims20