Matches in SemOpenAlex for { <https://semopenalex.org/work/W3209506041> ?p ?o ?g. }
Showing items 1 to 76 of
76
with 100 items per page.
- W3209506041 abstract "Adversarial Text Generation Frameworks (ATGFs) aim at causing a Natural Language Processing (NLP) machine to misbehave, i.e., misclassify a given input. In this paper, we propose EvilText, a general ATGF that successfully evades some of the most popular NLP machines by (efficiently) perturbing a given legitimate text, preserving at the same time the original text's semantics as well as human readability. Perturbations are based on visually similar classes of characters appearing in the unicode set. EvilText can be utilized from NLP services' operators for evaluating their systems security and robustness. Furthermore, EvilText outperforms the state-of-the-art ATGFs, in terms of: (a) effectiveness, (b) efficiency and (c) original text's semantics and human readability preservation. We evaluate EvilText on some of the most popular NLP systems used for sentiment analysis and toxic content detection. We further expand on the generality and transferability of our ATGF, while also exploring possible countermeasures for defending against our attacks. Surprisingly, naive defence mechanisms fail to mitigate our attacks; the only promising one being the restriction of unicode characters use. However, we argue that restricting the use of unicode characters imposes a significant trade-off between security and usability as almost all websites are heavily based on unicode support." @default.
- W3209506041 created "2021-11-08" @default.
- W3209506041 creator A5022960283 @default.
- W3209506041 creator A5047172111 @default.
- W3209506041 date "2021-11-15" @default.
- W3209506041 modified "2023-09-25" @default.
- W3209506041 title "Unicode Evil" @default.
- W3209506041 cites W1832693441 @default.
- W3209506041 cites W2151298633 @default.
- W3209506041 cites W2296629652 @default.
- W3209506041 cites W2799194071 @default.
- W3209506041 cites W2963859254 @default.
- W3209506041 cites W2963969878 @default.
- W3209506041 cites W3105389675 @default.
- W3209506041 cites W4205184193 @default.
- W3209506041 doi "https://doi.org/10.1145/3474369.3486871" @default.
- W3209506041 hasPublicationYear "2021" @default.
- W3209506041 type Work @default.
- W3209506041 sameAs 3209506041 @default.
- W3209506041 citedByCount "0" @default.
- W3209506041 crossrefType "proceedings-article" @default.
- W3209506041 hasAuthorship W3209506041A5022960283 @default.
- W3209506041 hasAuthorship W3209506041A5047172111 @default.
- W3209506041 hasBestOaLocation W32095060412 @default.
- W3209506041 hasConcept C104317684 @default.
- W3209506041 hasConcept C107457646 @default.
- W3209506041 hasConcept C154945302 @default.
- W3209506041 hasConcept C15744967 @default.
- W3209506041 hasConcept C170130773 @default.
- W3209506041 hasConcept C184337299 @default.
- W3209506041 hasConcept C185592680 @default.
- W3209506041 hasConcept C199360897 @default.
- W3209506041 hasConcept C204321447 @default.
- W3209506041 hasConcept C23123220 @default.
- W3209506041 hasConcept C2778143727 @default.
- W3209506041 hasConcept C2780767217 @default.
- W3209506041 hasConcept C41008148 @default.
- W3209506041 hasConcept C500551929 @default.
- W3209506041 hasConcept C542102704 @default.
- W3209506041 hasConcept C55493867 @default.
- W3209506041 hasConcept C63479239 @default.
- W3209506041 hasConceptScore W3209506041C104317684 @default.
- W3209506041 hasConceptScore W3209506041C107457646 @default.
- W3209506041 hasConceptScore W3209506041C154945302 @default.
- W3209506041 hasConceptScore W3209506041C15744967 @default.
- W3209506041 hasConceptScore W3209506041C170130773 @default.
- W3209506041 hasConceptScore W3209506041C184337299 @default.
- W3209506041 hasConceptScore W3209506041C185592680 @default.
- W3209506041 hasConceptScore W3209506041C199360897 @default.
- W3209506041 hasConceptScore W3209506041C204321447 @default.
- W3209506041 hasConceptScore W3209506041C23123220 @default.
- W3209506041 hasConceptScore W3209506041C2778143727 @default.
- W3209506041 hasConceptScore W3209506041C2780767217 @default.
- W3209506041 hasConceptScore W3209506041C41008148 @default.
- W3209506041 hasConceptScore W3209506041C500551929 @default.
- W3209506041 hasConceptScore W3209506041C542102704 @default.
- W3209506041 hasConceptScore W3209506041C55493867 @default.
- W3209506041 hasConceptScore W3209506041C63479239 @default.
- W3209506041 hasLocation W32095060411 @default.
- W3209506041 hasLocation W32095060412 @default.
- W3209506041 hasOpenAccess W3209506041 @default.
- W3209506041 hasPrimaryLocation W32095060411 @default.
- W3209506041 hasRelatedWork W1505979446 @default.
- W3209506041 hasRelatedWork W1541271503 @default.
- W3209506041 hasRelatedWork W1587224678 @default.
- W3209506041 hasRelatedWork W1855927527 @default.
- W3209506041 hasRelatedWork W2047793074 @default.
- W3209506041 hasRelatedWork W3009120927 @default.
- W3209506041 hasRelatedWork W3097853387 @default.
- W3209506041 hasRelatedWork W3214915308 @default.
- W3209506041 hasRelatedWork W4242922013 @default.
- W3209506041 hasRelatedWork W3202147931 @default.
- W3209506041 isParatext "false" @default.
- W3209506041 isRetracted "false" @default.
- W3209506041 magId "3209506041" @default.
- W3209506041 workType "article" @default.