Matches in SemOpenAlex for { <https://semopenalex.org/work/W3089357417> ?p ?o ?g. }
Showing items 1 to 100 of
100
with 100 items per page.
- W3089357417 abstract "Generating adversarial examples for natural language is hard, as natural language consists of discrete symbols, and examples are often of variable lengths. In this paper, we propose a geometry-inspired attack for generating natural language adversarial examples. Our attack generates adversarial examples by iteratively approximating the decision boundary of Deep Neural Networks (DNNs). Experiments on two datasets with two different models show that our attack fools natural language models with high success rates, while only replacing a few words. Human evaluation shows that adversarial examples generated by our attack are hard for humans to recognize. Further experiments show that adversarial training can improve model robustness against our attack." @default.
- W3089357417 created "2020-10-08" @default.
- W3089357417 creator A5066080136 @default.
- W3089357417 creator A5078339613 @default.
- W3089357417 date "2020-10-03" @default.
- W3089357417 modified "2023-09-23" @default.
- W3089357417 title "A Geometry-Inspired Attack for Generating Natural Language Adversarial Examples" @default.
- W3089357417 cites W1832693441 @default.
- W3089357417 cites W1932198206 @default.
- W3089357417 cites W2113459411 @default.
- W3089357417 cites W2194775991 @default.
- W3089357417 cites W2243397390 @default.
- W3089357417 cites W2250539671 @default.
- W3089357417 cites W2562979205 @default.
- W3089357417 cites W2799194071 @default.
- W3089357417 cites W2949128310 @default.
- W3089357417 cites W2950606982 @default.
- W3089357417 cites W2962700793 @default.
- W3089357417 cites W2962713901 @default.
- W3089357417 cites W2962718684 @default.
- W3089357417 cites W2962739339 @default.
- W3089357417 cites W2963026800 @default.
- W3089357417 cites W2963126845 @default.
- W3089357417 cites W2963207607 @default.
- W3089357417 cites W2963341956 @default.
- W3089357417 cites W2963748441 @default.
- W3089357417 cites W2963859254 @default.
- W3089357417 cites W2963969878 @default.
- W3089357417 cites W2964116600 @default.
- W3089357417 cites W2964152294 @default.
- W3089357417 cites W2964153729 @default.
- W3089357417 cites W2964159778 @default.
- W3089357417 cites W2964253222 @default.
- W3089357417 cites W2982756474 @default.
- W3089357417 doi "https://doi.org/10.48550/arxiv.2010.01345" @default.
- W3089357417 hasPublicationYear "2020" @default.
- W3089357417 type Work @default.
- W3089357417 sameAs 3089357417 @default.
- W3089357417 citedByCount "1" @default.
- W3089357417 countsByYear W30893574172021 @default.
- W3089357417 crossrefType "posted-content" @default.
- W3089357417 hasAuthorship W3089357417A5066080136 @default.
- W3089357417 hasAuthorship W3089357417A5078339613 @default.
- W3089357417 hasBestOaLocation W30893574171 @default.
- W3089357417 hasConcept C104317684 @default.
- W3089357417 hasConcept C134306372 @default.
- W3089357417 hasConcept C137293760 @default.
- W3089357417 hasConcept C154945302 @default.
- W3089357417 hasConcept C166957645 @default.
- W3089357417 hasConcept C185592680 @default.
- W3089357417 hasConcept C195324797 @default.
- W3089357417 hasConcept C205649164 @default.
- W3089357417 hasConcept C2776608160 @default.
- W3089357417 hasConcept C2779439875 @default.
- W3089357417 hasConcept C2984842247 @default.
- W3089357417 hasConcept C33923547 @default.
- W3089357417 hasConcept C37736160 @default.
- W3089357417 hasConcept C41008148 @default.
- W3089357417 hasConcept C50644808 @default.
- W3089357417 hasConcept C55493867 @default.
- W3089357417 hasConcept C62354387 @default.
- W3089357417 hasConcept C63479239 @default.
- W3089357417 hasConcept C80444323 @default.
- W3089357417 hasConceptScore W3089357417C104317684 @default.
- W3089357417 hasConceptScore W3089357417C134306372 @default.
- W3089357417 hasConceptScore W3089357417C137293760 @default.
- W3089357417 hasConceptScore W3089357417C154945302 @default.
- W3089357417 hasConceptScore W3089357417C166957645 @default.
- W3089357417 hasConceptScore W3089357417C185592680 @default.
- W3089357417 hasConceptScore W3089357417C195324797 @default.
- W3089357417 hasConceptScore W3089357417C205649164 @default.
- W3089357417 hasConceptScore W3089357417C2776608160 @default.
- W3089357417 hasConceptScore W3089357417C2779439875 @default.
- W3089357417 hasConceptScore W3089357417C2984842247 @default.
- W3089357417 hasConceptScore W3089357417C33923547 @default.
- W3089357417 hasConceptScore W3089357417C37736160 @default.
- W3089357417 hasConceptScore W3089357417C41008148 @default.
- W3089357417 hasConceptScore W3089357417C50644808 @default.
- W3089357417 hasConceptScore W3089357417C55493867 @default.
- W3089357417 hasConceptScore W3089357417C62354387 @default.
- W3089357417 hasConceptScore W3089357417C63479239 @default.
- W3089357417 hasConceptScore W3089357417C80444323 @default.
- W3089357417 hasLocation W30893574171 @default.
- W3089357417 hasLocation W30893574172 @default.
- W3089357417 hasOpenAccess W3089357417 @default.
- W3089357417 hasPrimaryLocation W30893574171 @default.
- W3089357417 hasRelatedWork W10234003 @default.
- W3089357417 hasRelatedWork W11990303 @default.
- W3089357417 hasRelatedWork W13032089 @default.
- W3089357417 hasRelatedWork W13972091 @default.
- W3089357417 hasRelatedWork W14719935 @default.
- W3089357417 hasRelatedWork W2085493 @default.
- W3089357417 hasRelatedWork W2867368 @default.
- W3089357417 hasRelatedWork W5179166 @default.
- W3089357417 hasRelatedWork W6518904 @default.
- W3089357417 hasRelatedWork W8053366 @default.
- W3089357417 isParatext "false" @default.
- W3089357417 isRetracted "false" @default.
- W3089357417 magId "3089357417" @default.
- W3089357417 workType "article" @default.