Dataset Search
Sort By
Search results
75 results found
Trojan Detection Software Challenge - llm-instruct-oct2024-test
Data provided by National Institute of Standards and Technology
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.
Tags: Trojan Detection; Artificial Intelligence; AI; Machine Learning; Adversarial Machine Learning;
Modified:
Source: https://drive.google.com/drive/folders/1BSGgEx56XWFG0ADOilZirPKbWkEHk3qF?usp=drive_link
Trojan Detection Software Challenge - rl-colorful-memory-sep2024-train
Data provided by National Institute of Standards and Technology
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.
Tags: Trojan Detection; Artificial Intelligence; AI; Machine Learning; Adversarial Machine Learning;
Modified:
Source: https://drive.google.com/drive/folders/1NUlgd1uy-GbbgdkXE_AEglC7tR3KZL7k?usp=sharing
Trojan Detection Software Challenge - rl-colorful-memory-sep2024-test
Data provided by National Institute of Standards and Technology
This is the test 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.
Tags: Trojan Detection; Artificial Intelligence; AI; Machine Learning; Adversarial Machine Learning;
Modified:
Source: https://drive.google.com/drive/folders/1ycEhNKR0M7lBrutifZonUiLf2uB-60Jt?usp=drive_link
Trojan Detection Software Challenge - rl-colorful-memory-sep2024-holdout
Data provided by National Institute of Standards and Technology
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.
Tags: Trojan Detection; Artificial Intelligence; AI; Machine Learning; Adversarial Machine Learning;
Modified:
Source: https://drive.google.com/drive/folders/1Oc_NJqPOi4BxQreCi1NestgqEeA8Vbtb?usp=drive_link
Trojan Detection Software Challenge - cyber-pe-aug2024-train
Data provided by National Institute of Standards and Technology
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.
Tags: Trojan Detection; Artificial Intelligence; AI; Machine Learning; Adversarial Machine Learning;
Modified:
Source: https://drive.google.com/file/d/1V28kBm3QR0lfk14RRxMMG0FMQCrOk2Ci/view?usp=drive_link
Trojan Detection Software Challenge - cyber-pe-aug2024-test
Data provided by National Institute of Standards and Technology
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.
Tags: Trojan Detection; Artificial Intelligence; AI; Machine Learning; Adversarial Machine Learning;
Modified:
Source: https://drive.google.com/drive/folders/1aUhJdOkMluaD539jonlcWSeF1zoB8hNr?usp=drive_link
Trojan Detection Software Challenge - cyber-pe-aug2024-holdout
Data provided by National Institute of Standards and Technology
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.
Tags: Trojan Detection; Artificial Intelligence; AI; Machine Learning; Adversarial Machine Learning;
Modified:
Source: https://drive.google.com/drive/folders/1G3RhnF5ini7Rtk0Pv0UfJLh_xGPFSRAq?usp=drive_link
Trojan Detection Software Challenge - mitigation-image-classification-jun2024-train
Data provided by National Institute of Standards and Technology
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.
Tags: Trojan Detection; Artificial Intelligence; AI; Machine Learning; Adversarial Machine Learning;
Modified:
Source: https://drive.google.com/drive/folders/1VJPQgyydbOifr0UXO2eybxZBN1a5uY6f?usp=sharing
Trojan Detection Software Challenge - mitigation-image-classification-jun2024-test
Data provided by National Institute of Standards and Technology
mitigation-image-classification-jun2024-test dataset
This 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.
Tags: Trojan Detection; Artificial Intelligence; AI; Machine Learning; Adversarial Machine Learning;
Modified:
Source: https://drive.google.com/drive/folders/1tj9AYRqRHA-vLBCcbXc-3T-D1LyE3vTb?usp=drive_link
Trojan Detection Software Challenge - llm-pretrain-apr2024-train
Data provided by National Institute of Standards and Technology
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.
Tags: Trojan Detection; Artificial Intelligence; AI; Machine Learning; Adversarial Machine Learning;
Modified:
Source: https://drive.google.com/drive/folders/1eI7MsVi1qqSHvnfCUWkgNnphTk0Cth5M?usp=sharing
