This collection of AI models consists of the trained yet unreleased models constructed for the TrojAI program. These models were not required for the published round they come from for some reason. Some were simply extras, others did not meet the release criteria for accuracy or trojan attack success rate, or other reasons. These models should not be considered high quality. They might be poorly trained and unfit for their stated purpose. Examine the metadata released with each model to determine whether you can leverage it.

Reference functional model for the activities and processes involved in production (e.g., designing, manufacturing, and recovering products/materials) in a circular economy. The model includes inputs and outputs of each activity, as well as factors that constrain and enable the activities (controls and mechanisms), using the IDEF0 methodology. It provides a baseline scenario that is relevant for stakeholders across product development and supply chain, including business strategists, material suppliers, designers, manufacturers, consumers, and recovers/recyclers.

This is the training data used to create and evaluate trojan detection software solutions. This data, generated at NIST, consists of RL agents operating in the Safety Gymnasium environment. A known percentage of these trained AI models have been poisoned with a known trigger which induces incorrect behavior. This data will be used to develop software solutions for detecting that trigger behavior in the trained AI models.

This is the training data used to create and evaluate trojan detection software solutions. This data, generated at NIST, consists of instruction fine tuned LLMs. A known percentage of these trained AI models have been poisoned with a known trigger which induces incorrect behavior. This data will be used to develop software solutions for mitigating that trigger behavior in the trained AI models.

This is the training data used to create and evaluate trojan detection software solutions. This data, generated at NIST, consists of instruction fine tuned LLMs. A known percentage of these trained AI models have been poisoned with a known trigger which induces incorrect behavior. This data will be used to develop software solutions for detecting that trigger behavior in the trained AI models.

This is the training data used to create and evaluate trojan detection software solutions. This data, generated at NIST, consists of models trained to predict whether code from public git repositories would survive in its branch for one month or more as a quantifiable proxy for code quality. A known percentage of these trained AI models have been poisoned with a known trigger which induces incorrect behavior. This data will be used to develop software solutions for detecting that trigger behavior in the trained AI models.

This is the training data used to create and evaluate trojan detection software solutions. This data, generated at NIST, consists of RL agents operating in the Colorful Memory environment. A known percentage of these trained AI models have been poisoned with a known trigger which induces incorrect behavior. This data will be used to develop software solutions for detecting that trigger behavior in the trained AI models.

This is the training data used to create and evaluate trojan detection software solutions. This data, generated at NIST, consists of malware packer classification AIs. A known percentage of these trained AI models have been poisoned with a known trigger which induces incorrect behavior. This data will be used to develop software solutions for mitigating/removing that trigger behavior from the trained AI models.

Sim-PROCESD is a discrete event simulation package written in Python that is designed to model the behavior of discrete manufacturing systems. Specifically, it focuses on asynchronous production lines. It also provides functionality for modeling the degradation and maintenance of machines in these systems. Sim-PROCESD provides class definitions for manufacturing devices/components that can be configured by the user to model various real-world manufacturing systems.

mitigation-image-classification-jun2024-train datasetThis is the training data used to create and evaluate trojan detection software solutions. This data, generated at NIST, consists of image classification AIs. A known percentage of these trained AI models have been poisoned with a known trigger which induces incorrect behavior. This data will be used to develop software solutions for mitigating/removing that trigger behavior from the trained AI models. This dataset consists of 288 AI models using a small set of model architectures.