Query-efficient model extraction for text classification model in a hard label setting

Query-efficient model extraction for text classification model in a hard label setting

Designing a query-efficient model extraction strategy to steal models from cloud-based platforms with black-box constraints remains a challenge, especially for language models. In a more realistic setting, a lack of information about the target model’s internal parameters, gradients, training data,...

Full description

Bibliographic Details
Main Authors: Hao Peng, Shixin Guo, Dandan Zhao, Yiming Wu, Jianming Han, Zhe Wang, Shouling Ji, Ming Zhong
Format: Article
Language:English
Published: Elsevier 2023-04-01
Series:Journal of King Saud University: Computer and Information Sciences
Subjects:
Model extraction
Language model stealing
Model privacy
Adversarial attack
Natural language processing
Performance Evaluation
Online Access:http://www.sciencedirect.com/science/article/pii/S1319157823000563

Internet

http://www.sciencedirect.com/science/article/pii/S1319157823000563

Similar Items