A cross-entropy optimization attack (CEA) with sememe- and MLM-based candidate words improves black-box adversarial attacks on classifiers and machine translation models.
In: Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, pp
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.CL 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
unclear 1representative citing papers
citing papers explorer
-
Cross-Entropy Attacks to Language Models via Rare Event Simulation
A cross-entropy optimization attack (CEA) with sememe- and MLM-based candidate words improves black-box adversarial attacks on classifiers and machine translation models.