TrojAI llm-pretrain-apr2024 Train DatasetThis is the training data used to create and evaluate trojan detection software solutions. This data, generated at NIST, consists Llama2 Large Language Models refined using fine-tuning and LoRA to perform next token prediction. A known percentage of these trained AI models have been poisoned with triggers which induces modified behavior. This data will be used to develop software solutions for detecting which trained AI models have been poisoned via embedded triggers into the model weights.

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.

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.

Supporting information for the NIST Internal Report entitled: Per- and Polyfluoroalkyl Substances - Non-Targeted Analysis Interlaboratory Study Final Report. The files include four comma-separated value (CSV) text files, three of the CSV text files contain summary tables of all per and polyfluoroalkyl substances (PFAS) detected in the three test samples from the interlaboratory report (sample_a_table.csv, sample_b_table.csv, sample_c_table.csv) including the individual laboratory reported identities and their associated identification confidence levels.

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 data publication contains the raw, extracted, and derived (e.g., peak areas) mass spectrometry data for illicit drug and mixture samples. These samples were analyzed by wipe-based collection or solvent-extraction, thermal desorption, and a transportable high-resolution time-of-flight mass spectrometer. Data were collected with both an in-line dielectric barrier discharge ionization source and an acetone-dopoant assisted vacuum ultraviolet (VUV) photoionization source.

These sets of automated serial sectioning electron backscatter diffraction (EBSD) and X-ray computed tomography (XRCT) measurements were designed to produce spatially registered, multi-modal datasets within a region of interest (ROI) of an additive manufacturing benchmark test series (AM Bench) nickel alloy 718 specimen produced using laser powder bed fusion (PBF-LB) additive manufacturing (AM). The serial sectioning data are in the form of 2D slices which were reconstructed and registered to make 3D data sets that can be analyzed and visualized using open source software.

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.