Matches in SemOpenAlex for { <https://semopenalex.org/work/W3161994961> ?p ?o ?g. }
- W3161994961 abstract "We describe models focused at the understudied problem of translating between monolingual and code-mixed language pairs. More specifically, we offer a wide range of models that convert monolingual English text into Hinglish (code-mixed Hindi and English). Given the recent success of pretrained language models, we also test the utility of two recent Transformer-based encoder-decoder models (i.e., mT5 and mBART) on the task finding both to work well. Given the paucity of training data for code-mixing, we also propose a dependency-free method for generating code-mixed texts from bilingual distributed representations that we exploit for improving language model performance. In particular, armed with this additional data, we adopt a curriculum learning approach where we first finetune the language models on synthetic data then on gold code-mixed data. We find that, although simple, our synthetic code-mixing method is competitive with (and in some cases is even superior to) several standard methods (backtranslation, method based on equivalence constraint theory) under a diverse set of conditions. Our work shows that the mT5 model, finetuned following the curriculum learning procedure, achieves best translation performance (12.67 BLEU). Our models place first in the overall ranking of the English-Hinglish official shared task." @default.
- W3161994961 created "2021-05-24" @default.
- W3161994961 creator A5004629670 @default.
- W3161994961 creator A5041553790 @default.
- W3161994961 creator A5057854225 @default.
- W3161994961 creator A5061340195 @default.
- W3161994961 date "2021-05-18" @default.
- W3161994961 modified "2023-09-27" @default.
- W3161994961 title "Exploring Text-to-Text Transformers for English to Hinglish Machine Translation with Synthetic Code-Mixing" @default.
- W3161994961 cites W1527783480 @default.
- W3161994961 cites W2014962660 @default.
- W3161994961 cites W2061272101 @default.
- W3161994961 cites W2159849140 @default.
- W3161994961 cites W2798348125 @default.
- W3161994961 cites W2807007025 @default.
- W3161994961 cites W2807157666 @default.
- W3161994961 cites W2928484296 @default.
- W3161994961 cites W2963341956 @default.
- W3161994961 cites W2963508788 @default.
- W3161994961 cites W2963877297 @default.
- W3161994961 cites W2963945964 @default.
- W3161994961 cites W2964327384 @default.
- W3161994961 cites W2966263504 @default.
- W3161994961 cites W2983040767 @default.
- W3161994961 cites W2989143494 @default.
- W3161994961 cites W2996287690 @default.
- W3161994961 cites W3082274269 @default.
- W3161994961 cites W3088665616 @default.
- W3161994961 cites W3098670961 @default.
- W3161994961 cites W3098824823 @default.
- W3161994961 cites W3103037747 @default.
- W3161994961 cites W3107826490 @default.
- W3161994961 cites W3153869297 @default.
- W3161994961 cites W3169483174 @default.
- W3161994961 hasPublicationYear "2021" @default.
- W3161994961 type Work @default.
- W3161994961 sameAs 3161994961 @default.
- W3161994961 citedByCount "0" @default.
- W3161994961 crossrefType "posted-content" @default.
- W3161994961 hasAuthorship W3161994961A5004629670 @default.
- W3161994961 hasAuthorship W3161994961A5041553790 @default.
- W3161994961 hasAuthorship W3161994961A5057854225 @default.
- W3161994961 hasAuthorship W3161994961A5061340195 @default.
- W3161994961 hasConcept C111919701 @default.
- W3161994961 hasConcept C118505674 @default.
- W3161994961 hasConcept C121332964 @default.
- W3161994961 hasConcept C137293760 @default.
- W3161994961 hasConcept C138885662 @default.
- W3161994961 hasConcept C154945302 @default.
- W3161994961 hasConcept C165696696 @default.
- W3161994961 hasConcept C165801399 @default.
- W3161994961 hasConcept C177264268 @default.
- W3161994961 hasConcept C18552078 @default.
- W3161994961 hasConcept C19768560 @default.
- W3161994961 hasConcept C199360897 @default.
- W3161994961 hasConcept C203005215 @default.
- W3161994961 hasConcept C204321447 @default.
- W3161994961 hasConcept C2776760102 @default.
- W3161994961 hasConcept C38652104 @default.
- W3161994961 hasConcept C41008148 @default.
- W3161994961 hasConcept C41895202 @default.
- W3161994961 hasConcept C62520636 @default.
- W3161994961 hasConcept C66322947 @default.
- W3161994961 hasConceptScore W3161994961C111919701 @default.
- W3161994961 hasConceptScore W3161994961C118505674 @default.
- W3161994961 hasConceptScore W3161994961C121332964 @default.
- W3161994961 hasConceptScore W3161994961C137293760 @default.
- W3161994961 hasConceptScore W3161994961C138885662 @default.
- W3161994961 hasConceptScore W3161994961C154945302 @default.
- W3161994961 hasConceptScore W3161994961C165696696 @default.
- W3161994961 hasConceptScore W3161994961C165801399 @default.
- W3161994961 hasConceptScore W3161994961C177264268 @default.
- W3161994961 hasConceptScore W3161994961C18552078 @default.
- W3161994961 hasConceptScore W3161994961C19768560 @default.
- W3161994961 hasConceptScore W3161994961C199360897 @default.
- W3161994961 hasConceptScore W3161994961C203005215 @default.
- W3161994961 hasConceptScore W3161994961C204321447 @default.
- W3161994961 hasConceptScore W3161994961C2776760102 @default.
- W3161994961 hasConceptScore W3161994961C38652104 @default.
- W3161994961 hasConceptScore W3161994961C41008148 @default.
- W3161994961 hasConceptScore W3161994961C41895202 @default.
- W3161994961 hasConceptScore W3161994961C62520636 @default.
- W3161994961 hasConceptScore W3161994961C66322947 @default.
- W3161994961 hasLocation W31619949611 @default.
- W3161994961 hasOpenAccess W3161994961 @default.
- W3161994961 hasPrimaryLocation W31619949611 @default.
- W3161994961 hasRelatedWork W2181660330 @default.
- W3161994961 hasRelatedWork W2186342161 @default.
- W3161994961 hasRelatedWork W2187622083 @default.
- W3161994961 hasRelatedWork W2408503330 @default.
- W3161994961 hasRelatedWork W2481351509 @default.
- W3161994961 hasRelatedWork W2596454883 @default.
- W3161994961 hasRelatedWork W2956885163 @default.
- W3161994961 hasRelatedWork W2988936111 @default.
- W3161994961 hasRelatedWork W3014391559 @default.
- W3161994961 hasRelatedWork W3022839720 @default.
- W3161994961 hasRelatedWork W3093517588 @default.
- W3161994961 hasRelatedWork W3103037747 @default.
- W3161994961 hasRelatedWork W3118612332 @default.
- W3161994961 hasRelatedWork W3119866316 @default.