Public Data Resource

Trojan Detection Software Challenge - nlp-summary-jan2022-holdout

Contact: Michael Paul Majurski..
Identifier: doi:10.18434/mds2-2782
Version: 1.0...


Round 9 Holdout Dataset

This is the holdout data used to evaluate trojan detection software solutions. This data, generated at NIST, consists of natural language processing (NLP) AIs trained to perform one of three tasks, sentiment classification, named entity recognition, or extractive question answering on English text. 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 which trained AI models have been poisoned via embedded triggers. This dataset consists of 410 Sentiment Classification, Named Entity Recognition, and Extractive Question Answering AI models using a small set of model architectures. Half (50%) of the models have been poisoned with an embedded trigger which causes misclassification of the input when the trigger is present.
Research Topics: Information Technology: Software research, Information Technology: Cybersecurity    
Subject Keywords: Trojan Detection; Artificial Intelligence; AI; Machine Learning; Adversarial Machine Learning;    

Data Access

These data are public.
Data and related material can be found at the following locations:

About This Dataset

Version: 1.0...
Cite this dataset
Michael Paul Majurski (2022), Trojan Detection Software Challenge - nlp-summary-jan2022-holdout, National Institute of Standards and Technology, (Accessed 2024-07-15)
Repository Metadata
Machine-readable descriptions of this dataset are available in the following formats:
Access Metrics
Metrics data is not available for all datasets, including this one. This may be because the data is served via servers external to this repository.