Matches in SemOpenAlex for { <https://semopenalex.org/work/W3205088081> ?p ?o ?g. }
- W3205088081 abstract "Recent advancements in NLP have given us models like mBERT and XLMR that can serve over 100 languages. The languages that these models are evaluated on, however, are very few in number, and it is unlikely that evaluation datasets will cover all the languages that these models support. Potential solutions to the costly problem of dataset creation are to translate datasets to new languages or use template-filling based techniques for creation. This paper proposes an alternate solution for evaluating a model across languages which make use of the existing performance scores of the model on languages that a particular task has test sets for. We train a predictor on these performance scores and use this predictor to predict the model's performance in different evaluation settings. Our results show that our method is effective in filling the gaps in the evaluation for an existing set of languages, but might require additional improvements if we want it to generalize to unseen languages." @default.
- W3205088081 created "2021-10-25" @default.
- W3205088081 creator A5005513786 @default.
- W3205088081 creator A5006779728 @default.
- W3205088081 creator A5008944385 @default.
- W3205088081 creator A5013639574 @default.
- W3205088081 creator A5024685185 @default.
- W3205088081 creator A5045216067 @default.
- W3205088081 date "2021-10-17" @default.
- W3205088081 modified "2023-10-14" @default.
- W3205088081 title "Predicting the Performance of Multilingual NLP Models." @default.
- W3205088081 cites W2079656678 @default.
- W3205088081 cites W2550821151 @default.
- W3205088081 cites W2739967986 @default.
- W3205088081 cites W2752630748 @default.
- W3205088081 cites W2890225082 @default.
- W3205088081 cites W2896457183 @default.
- W3205088081 cites W2914120296 @default.
- W3205088081 cites W2947992883 @default.
- W3205088081 cites W2948384082 @default.
- W3205088081 cites W2953109491 @default.
- W3205088081 cites W2970413168 @default.
- W3205088081 cites W2971863715 @default.
- W3205088081 cites W2995015695 @default.
- W3205088081 cites W2995647371 @default.
- W3205088081 cites W2996580882 @default.
- W3205088081 cites W3013840636 @default.
- W3205088081 cites W3019003205 @default.
- W3205088081 cites W3021240322 @default.
- W3205088081 cites W3021774381 @default.
- W3205088081 cites W3027825632 @default.
- W3205088081 cites W3033940819 @default.
- W3205088081 cites W3034469191 @default.
- W3205088081 cites W3034776473 @default.
- W3205088081 cites W3035390927 @default.
- W3205088081 cites W3035507081 @default.
- W3205088081 cites W3035579820 @default.
- W3205088081 cites W3045462440 @default.
- W3205088081 cites W3093517588 @default.
- W3205088081 cites W3100198908 @default.
- W3205088081 cites W3127454544 @default.
- W3205088081 cites W3176765167 @default.
- W3205088081 hasPublicationYear "2021" @default.
- W3205088081 type Work @default.
- W3205088081 sameAs 3205088081 @default.
- W3205088081 citedByCount "0" @default.
- W3205088081 crossrefType "posted-content" @default.
- W3205088081 hasAuthorship W3205088081A5005513786 @default.
- W3205088081 hasAuthorship W3205088081A5006779728 @default.
- W3205088081 hasAuthorship W3205088081A5008944385 @default.
- W3205088081 hasAuthorship W3205088081A5013639574 @default.
- W3205088081 hasAuthorship W3205088081A5024685185 @default.
- W3205088081 hasAuthorship W3205088081A5045216067 @default.
- W3205088081 hasConcept C119857082 @default.
- W3205088081 hasConcept C137293760 @default.
- W3205088081 hasConcept C154945302 @default.
- W3205088081 hasConcept C162324750 @default.
- W3205088081 hasConcept C169903167 @default.
- W3205088081 hasConcept C177264268 @default.
- W3205088081 hasConcept C187736073 @default.
- W3205088081 hasConcept C199360897 @default.
- W3205088081 hasConcept C204321447 @default.
- W3205088081 hasConcept C2780451532 @default.
- W3205088081 hasConcept C41008148 @default.
- W3205088081 hasConceptScore W3205088081C119857082 @default.
- W3205088081 hasConceptScore W3205088081C137293760 @default.
- W3205088081 hasConceptScore W3205088081C154945302 @default.
- W3205088081 hasConceptScore W3205088081C162324750 @default.
- W3205088081 hasConceptScore W3205088081C169903167 @default.
- W3205088081 hasConceptScore W3205088081C177264268 @default.
- W3205088081 hasConceptScore W3205088081C187736073 @default.
- W3205088081 hasConceptScore W3205088081C199360897 @default.
- W3205088081 hasConceptScore W3205088081C204321447 @default.
- W3205088081 hasConceptScore W3205088081C2780451532 @default.
- W3205088081 hasConceptScore W3205088081C41008148 @default.
- W3205088081 hasLocation W32050880811 @default.
- W3205088081 hasOpenAccess W3205088081 @default.
- W3205088081 hasPrimaryLocation W32050880811 @default.
- W3205088081 hasRelatedWork W2945542139 @default.
- W3205088081 hasRelatedWork W2964000434 @default.
- W3205088081 hasRelatedWork W2970457158 @default.
- W3205088081 hasRelatedWork W2971045469 @default.
- W3205088081 hasRelatedWork W2988346595 @default.
- W3205088081 hasRelatedWork W3011654523 @default.
- W3205088081 hasRelatedWork W3016074070 @default.
- W3205088081 hasRelatedWork W3081145646 @default.
- W3205088081 hasRelatedWork W3094999800 @default.
- W3205088081 hasRelatedWork W3115280218 @default.
- W3205088081 hasRelatedWork W3126739176 @default.
- W3205088081 hasRelatedWork W3172076499 @default.
- W3205088081 hasRelatedWork W3179706568 @default.
- W3205088081 hasRelatedWork W3185555427 @default.
- W3205088081 hasRelatedWork W3201378203 @default.
- W3205088081 hasRelatedWork W3202954083 @default.
- W3205088081 hasRelatedWork W3204607391 @default.
- W3205088081 hasRelatedWork W3207893443 @default.
- W3205088081 hasRelatedWork W3209222664 @default.
- W3205088081 hasRelatedWork W2799362914 @default.
- W3205088081 isParatext "false" @default.
- W3205088081 isRetracted "false" @default.