3 Hits in 5.5 sec

Identifying Adversarial Attacks on Text Classifiers [article]

Zhouhang Xie, Jonathan Brophy, Adam Noack, Wencong You, Kalyani Asthana, Carter Perkins, Sabrina Reis, Sameer Singh, Daniel Lowd
2022 arXiv   pre-print
The landscape of adversarial attacks against text classifiers continues to grow, with new attacks developed every year and many of them available in standard toolkits, such as TextAttack and OpenAttack  ...  As a third contribution, we demonstrate the effectiveness of three classes of features for these tasks: text properties, capturing content and presentation of text; language model properties, determining  ...  This work benefited from access to the University of Oregon high performance computer, Talapas.  ... 
arXiv:2201.08555v1 fatcat:bknr7chhaza2bhnrwveufhot2m

Certified Robustness to Programmable Transformations in LSTMs

Yuhao Zhang, Aws Albarghouthi, Loris D'Antoni
2021 Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing   unpublished
NDSS 2019, San Diego, California, USA, February Vechev. 2018. AI2: safety and robustness certifi- 24-27, 2019. cation of neural networks with abstract interpretation.  ...  In 26th Annual Timon Gehr, Matthew Mirman, Dana Drachsler-Cohen, Network and Distributed System Security Symposium, Petar Tsankov, Swarat Chaudhuri, and Martin T.  ... 
doi:10.18653/v1/2021.emnlp-main.82 fatcat:zntt7jy6drcadnorq6kkwp64iy

Dynamic iOS Privacy Analysis: Verifying App Store Privacy Labels

Thomas Jirout, Martina Lindorfer
One major shortcoming of the system is that these declarations are not verified or audited by Apple, opening the possibility for potential misuse.The aim of this thesis was to evaluate how dynamic mobile  ...  They require iOS apps to disclose which kinds of data types they or their third-party partners collect and how they store and use it.  ...  In Proceedings of the Network and Distributed System Security Symposium, NDSS 2011, San Diego, California, USA, 6th February - 9th February 2011.  ... 
doi:10.34726/hss.2021.92880 fatcat:xeqh236swnfivk2p62uow3nmpq