Matches in SemOpenAlex for { <https://semopenalex.org/work/W3130812595> ?p ?o ?g. }
- W3130812595 abstract "In recent studies, it has been shown that Multilingual language models underperform their monolingual counterparts. It is also a well-known fact that training and maintaining monolingual models for each language is a costly and time-consuming process. Roman Urdu is a resource-starved language used popularly on social media platforms and chat apps. In this research, we propose a novel dataset of scraped tweets containing 54M tokens and 3M sentences. Additionally, we also propose RUBERT a bilingual Roman Urdu model created by additional pretraining of English BERT. We compare its performance with a monolingual Roman Urdu BERT trained from scratch and a multilingual Roman Urdu BERT created by additional pretraining of Multilingual BERT. We show through our experiments that additional pretraining of the English BERT produces the most notable performance improvement." @default.
- W3130812595 created "2021-03-01" @default.
- W3130812595 creator A5026645452 @default.
- W3130812595 creator A5057413825 @default.
- W3130812595 creator A5061521072 @default.
- W3130812595 date "2021-02-22" @default.
- W3130812595 modified "2023-09-27" @default.
- W3130812595 title "RUBERT: A Bilingual Roman Urdu BERT Using Cross Lingual Transfer Learning." @default.
- W3130812595 cites W2013962395 @default.
- W3130812595 cites W2060172308 @default.
- W3130812595 cites W2114359199 @default.
- W3130812595 cites W2115222200 @default.
- W3130812595 cites W2117130368 @default.
- W3130812595 cites W2118788272 @default.
- W3130812595 cites W2132339004 @default.
- W3130812595 cites W2153579005 @default.
- W3130812595 cites W2161918995 @default.
- W3130812595 cites W2171190214 @default.
- W3130812595 cites W2182195832 @default.
- W3130812595 cites W2250539671 @default.
- W3130812595 cites W2339779833 @default.
- W3130812595 cites W2402630038 @default.
- W3130812595 cites W2508023093 @default.
- W3130812595 cites W2766020882 @default.
- W3130812595 cites W2786363455 @default.
- W3130812595 cites W2792502426 @default.
- W3130812595 cites W2883300051 @default.
- W3130812595 cites W2890842468 @default.
- W3130812595 cites W2901949782 @default.
- W3130812595 cites W2909597359 @default.
- W3130812595 cites W2910987184 @default.
- W3130812595 cites W2945998518 @default.
- W3130812595 cites W2954996676 @default.
- W3130812595 cites W2962739339 @default.
- W3130812595 cites W2963026768 @default.
- W3130812595 cites W2963159690 @default.
- W3130812595 cites W2963310665 @default.
- W3130812595 cites W2963323070 @default.
- W3130812595 cites W2963341956 @default.
- W3130812595 cites W2963403868 @default.
- W3130812595 cites W2963626623 @default.
- W3130812595 cites W2963748441 @default.
- W3130812595 cites W2965373594 @default.
- W3130812595 cites W2970597249 @default.
- W3130812595 cites W2988923581 @default.
- W3130812595 cites W2995647371 @default.
- W3130812595 cites W3001112740 @default.
- W3130812595 cites W3003878885 @default.
- W3130812595 cites W3005736358 @default.
- W3130812595 cites W3005923092 @default.
- W3130812595 cites W3012385961 @default.
- W3130812595 cites W3033940819 @default.
- W3130812595 cites W3035390927 @default.
- W3130812595 cites W3036524591 @default.
- W3130812595 cites W3036840722 @default.
- W3130812595 cites W3085802240 @default.
- W3130812595 cites W3122276068 @default.
- W3130812595 cites W3126005225 @default.
- W3130812595 cites W2914696592 @default.
- W3130812595 hasPublicationYear "2021" @default.
- W3130812595 type Work @default.
- W3130812595 sameAs 3130812595 @default.
- W3130812595 citedByCount "0" @default.
- W3130812595 crossrefType "posted-content" @default.
- W3130812595 hasAuthorship W3130812595A5026645452 @default.
- W3130812595 hasAuthorship W3130812595A5057413825 @default.
- W3130812595 hasAuthorship W3130812595A5061521072 @default.
- W3130812595 hasConcept C138885662 @default.
- W3130812595 hasConcept C154945302 @default.
- W3130812595 hasConcept C204321447 @default.
- W3130812595 hasConcept C2777350258 @default.
- W3130812595 hasConcept C2780035574 @default.
- W3130812595 hasConcept C41008148 @default.
- W3130812595 hasConcept C41895202 @default.
- W3130812595 hasConceptScore W3130812595C138885662 @default.
- W3130812595 hasConceptScore W3130812595C154945302 @default.
- W3130812595 hasConceptScore W3130812595C204321447 @default.
- W3130812595 hasConceptScore W3130812595C2777350258 @default.
- W3130812595 hasConceptScore W3130812595C2780035574 @default.
- W3130812595 hasConceptScore W3130812595C41008148 @default.
- W3130812595 hasConceptScore W3130812595C41895202 @default.
- W3130812595 hasLocation W31308125951 @default.
- W3130812595 hasOpenAccess W3130812595 @default.
- W3130812595 hasPrimaryLocation W31308125951 @default.
- W3130812595 hasRelatedWork W2252113950 @default.
- W3130812595 hasRelatedWork W2973071945 @default.
- W3130812595 hasRelatedWork W3012990076 @default.
- W3130812595 hasRelatedWork W3025740135 @default.
- W3130812595 hasRelatedWork W3035262097 @default.
- W3130812595 hasRelatedWork W3080175947 @default.
- W3130812595 hasRelatedWork W3082873023 @default.
- W3130812595 hasRelatedWork W3088262928 @default.
- W3130812595 hasRelatedWork W3094458375 @default.
- W3130812595 hasRelatedWork W3101486370 @default.
- W3130812595 hasRelatedWork W3101860695 @default.
- W3130812595 hasRelatedWork W3113914437 @default.
- W3130812595 hasRelatedWork W3114757058 @default.
- W3130812595 hasRelatedWork W3120253119 @default.
- W3130812595 hasRelatedWork W3129648567 @default.
- W3130812595 hasRelatedWork W3136221257 @default.