FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3125752017> ?p ?o ?g. }

• W3125752017 abstract "Text style transfer has gained increasing attention from the research community over the recent years. However, the proposed approaches vary in many ways, which makes it hard to assess the individual contribution of the model components. In style transfer, the most important component is the optimization technique used to guide the learning in the absence of parallel training data. In this work we empirically compare the dominant optimization paradigms which provide supervision signals during training: backtranslation, adversarial training and reinforcement learning. We find that backtranslation has model-specific limitations, which inhibits training style transfer models. Reinforcement learning shows the best performance gains, while adversarial training, despite its popularity, does not offer an advantage over the latter alternative. In this work we also experiment with Minimum Risk Training, a popular technique in the machine translation community, which, to our knowledge, has not been empirically evaluated in the task of style transfer. We fill this research gap and empirically show its efficacy." @default.
• W3125752017 created "2021-02-01" @default.
• W3125752017 creator A5007087783 @default.
• W3125752017 creator A5027062608 @default.
• W3125752017 creator A5027450194 @default.
• W3125752017 creator A5028614494 @default.
• W3125752017 date "2021-01-19" @default.
• W3125752017 modified "2023-09-27" @default.
• W3125752017 title "Empirical Evaluation of Supervision Signals for Style Transfer Models" @default.
• W3125752017 cites W1498436455 @default.
• W3125752017 cites W1559169059 @default.
• W3125752017 cites W1614298861 @default.
• W3125752017 cites W168564468 @default.
• W3125752017 cites W2075201173 @default.
• W3125752017 cites W2099471712 @default.
• W3125752017 cites W2101105183 @default.
• W3125752017 cites W2125101937 @default.
• W3125752017 cites W2134800885 @default.
• W3125752017 cites W2143017621 @default.
• W3125752017 cites W2146574666 @default.
• W3125752017 cites W2157331557 @default.
• W3125752017 cites W2250445771 @default.
• W3125752017 cites W2250616809 @default.
• W3125752017 cites W2468328197 @default.
• W3125752017 cites W2539033431 @default.
• W3125752017 cites W2539350388 @default.
• W3125752017 cites W2617573776 @default.
• W3125752017 cites W2798273548 @default.
• W3125752017 cites W2888161220 @default.
• W3125752017 cites W2891348164 @default.
• W3125752017 cites W2914442349 @default.
• W3125752017 cites W2925000324 @default.
• W3125752017 cites W2927085091 @default.
• W3125752017 cites W2933374552 @default.
• W3125752017 cites W2947054491 @default.
• W3125752017 cites W2952335829 @default.
• W3125752017 cites W2962686347 @default.
• W3125752017 cites W2962788902 @default.
• W3125752017 cites W2962917899 @default.
• W3125752017 cites W2962937198 @default.
• W3125752017 cites W2963216553 @default.
• W3125752017 cites W2963366196 @default.
• W3125752017 cites W2963403868 @default.
• W3125752017 cites W2963463964 @default.
• W3125752017 cites W2963631950 @default.
• W3125752017 cites W2963667126 @default.
• W3125752017 cites W2964008635 @default.
• W3125752017 cites W2964121744 @default.
• W3125752017 cites W2964201867 @default.
• W3125752017 cites W2964308564 @default.
• W3125752017 cites W2964321064 @default.
• W3125752017 cites W2970441781 @default.
• W3125752017 cites W2970948593 @default.
• W3125752017 cites W3035125262 @default.
• W3125752017 hasPublicationYear "2021" @default.
• W3125752017 type Work @default.
• W3125752017 sameAs 3125752017 @default.
• W3125752017 citedByCount "0" @default.
• W3125752017 crossrefType "posted-content" @default.
• W3125752017 hasAuthorship W3125752017A5007087783 @default.
• W3125752017 hasAuthorship W3125752017A5027062608 @default.
• W3125752017 hasAuthorship W3125752017A5027450194 @default.
• W3125752017 hasAuthorship W3125752017A5028614494 @default.
• W3125752017 hasConcept C119857082 @default.
• W3125752017 hasConcept C121332964 @default.
• W3125752017 hasConcept C127413603 @default.
• W3125752017 hasConcept C150899416 @default.
• W3125752017 hasConcept C153294291 @default.
• W3125752017 hasConcept C154945302 @default.
• W3125752017 hasConcept C15744967 @default.
• W3125752017 hasConcept C162324750 @default.
• W3125752017 hasConcept C166957645 @default.
• W3125752017 hasConcept C173608175 @default.
• W3125752017 hasConcept C18762648 @default.
• W3125752017 hasConcept C187736073 @default.
• W3125752017 hasConcept C2776175482 @default.
• W3125752017 hasConcept C2776445246 @default.
• W3125752017 hasConcept C2777211547 @default.
• W3125752017 hasConcept C2777938197 @default.
• W3125752017 hasConcept C2780451532 @default.
• W3125752017 hasConcept C2780586970 @default.
• W3125752017 hasConcept C37736160 @default.
• W3125752017 hasConcept C41008148 @default.
• W3125752017 hasConcept C56739046 @default.
• W3125752017 hasConcept C77805123 @default.
• W3125752017 hasConcept C78519656 @default.
• W3125752017 hasConcept C95457728 @default.
• W3125752017 hasConcept C97541855 @default.
• W3125752017 hasConceptScore W3125752017C119857082 @default.
• W3125752017 hasConceptScore W3125752017C121332964 @default.
• W3125752017 hasConceptScore W3125752017C127413603 @default.
• W3125752017 hasConceptScore W3125752017C150899416 @default.
• W3125752017 hasConceptScore W3125752017C153294291 @default.
• W3125752017 hasConceptScore W3125752017C154945302 @default.
• W3125752017 hasConceptScore W3125752017C15744967 @default.
• W3125752017 hasConceptScore W3125752017C162324750 @default.
• W3125752017 hasConceptScore W3125752017C166957645 @default.
• W3125752017 hasConceptScore W3125752017C173608175 @default.
• W3125752017 hasConceptScore W3125752017C18762648 @default.
• W3125752017 hasConceptScore W3125752017C187736073 @default.

• next