Cross-model format comparison patent

ChangeBridge: Patent Grants - AI & Computing (G06N)

Published April 7th, 2026

Detected April 7th, 2026

Summary

USPTO granted HiddenLayer, Inc. Patent US12596965B2 covering methods for comparing machine learning model formats by extracting layer information and matching against known models to determine model identity. The patent includes 25 claims and covers apparatus, systems, and techniques for ML model comparison and identification.

View original document View source feed page

What changed

USPTO issued Patent US12596965B2 to HiddenLayer, Inc. for methods of obtaining machine learning model representations, extracting layer information, and comparing against known models to determine matches. The patent covers apparatus, systems, and techniques for automated ML model identification based on structural characteristics.

For technology companies developing or deploying machine learning models, this patent establishes proprietary rights in model format comparison technology. While patent grants do not create immediate compliance obligations, entities operating in AI/ML model analysis, security, or reverse engineering spaces should be aware of this intellectual property when developing similar capabilities.

What to do next

Monitor for updates

Source document (simplified)

← USPTO Patent Grants

Cross-model format comparison

Grant US12596965B2 Kind: B2 Apr 07, 2026

Assignee

HiddenLayer, Inc.

Inventors

Kasimir Ralph Peter Schulz, David Beveridge

Abstract

A machine learning model representation is obtained from a model source and information characterizing the layers of the model representation is extracted to result in extracted model information. This extracted model information can be compared to information characterizing one or more known (i.e., previously characterized) machine learning models in order to determine whether there is a match based on layer information. A match can, in some cases, be used to determine an identity of the underlying machine learning model for the model representation. Information regarding the comparison (i.e., the model matching determination) can be provided to a consuming application or process. Related apparatus, systems, techniques and articles are also described.