Matches in SemOpenAlex for { <https://semopenalex.org/work/W4385573307> ?p ?o ?g. }
Showing items 1 to 55 of
55
with 100 items per page.
- W4385573307 abstract "In this paper, we look at the case of a Generic text-to-text NMT model that has to deal with data coming from various modalities, like speech, images, or noisy text extracted from the web. We propose a two-step method, based on composable adapters, to deal with this problem of Multimodal Robustness. In a first step, we separately learn domain adapters and modality specific adapters, to deal with noisy input coming from various sources: ASR, OCR, or noisy text (UGC). In a second step, we combine these components at runtime via dynamic routing or, when the source of noise is unknown, via two new transfer learning mechanisms (Fast Fusion and Multi Fusion). We show that our method provides a flexible, state-of-the-art, architecture able to deal with noisy multimodal inputs." @default.
- W4385573307 created "2023-08-05" @default.
- W4385573307 creator A5034647649 @default.
- W4385573307 creator A5041255737 @default.
- W4385573307 date "2022-01-01" @default.
- W4385573307 modified "2023-09-24" @default.
- W4385573307 title "Multimodal Robustness for Neural Machine Translation" @default.
- W4385573307 doi "https://doi.org/10.18653/v1/2022.emnlp-main.582" @default.
- W4385573307 hasPublicationYear "2022" @default.
- W4385573307 type Work @default.
- W4385573307 citedByCount "0" @default.
- W4385573307 crossrefType "proceedings-article" @default.
- W4385573307 hasAuthorship W4385573307A5034647649 @default.
- W4385573307 hasAuthorship W4385573307A5041255737 @default.
- W4385573307 hasBestOaLocation W43855733071 @default.
- W4385573307 hasConcept C104317684 @default.
- W4385573307 hasConcept C119857082 @default.
- W4385573307 hasConcept C144024400 @default.
- W4385573307 hasConcept C154945302 @default.
- W4385573307 hasConcept C185592680 @default.
- W4385573307 hasConcept C203005215 @default.
- W4385573307 hasConcept C2779903281 @default.
- W4385573307 hasConcept C28490314 @default.
- W4385573307 hasConcept C36289849 @default.
- W4385573307 hasConcept C41008148 @default.
- W4385573307 hasConcept C55493867 @default.
- W4385573307 hasConcept C63479239 @default.
- W4385573307 hasConceptScore W4385573307C104317684 @default.
- W4385573307 hasConceptScore W4385573307C119857082 @default.
- W4385573307 hasConceptScore W4385573307C144024400 @default.
- W4385573307 hasConceptScore W4385573307C154945302 @default.
- W4385573307 hasConceptScore W4385573307C185592680 @default.
- W4385573307 hasConceptScore W4385573307C203005215 @default.
- W4385573307 hasConceptScore W4385573307C2779903281 @default.
- W4385573307 hasConceptScore W4385573307C28490314 @default.
- W4385573307 hasConceptScore W4385573307C36289849 @default.
- W4385573307 hasConceptScore W4385573307C41008148 @default.
- W4385573307 hasConceptScore W4385573307C55493867 @default.
- W4385573307 hasConceptScore W4385573307C63479239 @default.
- W4385573307 hasLocation W43855733071 @default.
- W4385573307 hasOpenAccess W4385573307 @default.
- W4385573307 hasPrimaryLocation W43855733071 @default.
- W4385573307 hasRelatedWork W1974878518 @default.
- W4385573307 hasRelatedWork W2016948484 @default.
- W4385573307 hasRelatedWork W2392593357 @default.
- W4385573307 hasRelatedWork W2435130738 @default.
- W4385573307 hasRelatedWork W2961085424 @default.
- W4385573307 hasRelatedWork W3034443970 @default.
- W4385573307 hasRelatedWork W3140942413 @default.
- W4385573307 hasRelatedWork W4286629047 @default.
- W4385573307 hasRelatedWork W4306674287 @default.
- W4385573307 hasRelatedWork W4224009465 @default.
- W4385573307 isParatext "false" @default.
- W4385573307 isRetracted "false" @default.
- W4385573307 workType "article" @default.