Matches in SemOpenAlex for { <https://semopenalex.org/work/W4385572734> ?p ?o ?g. }
Showing items 1 to 63 of
63
with 100 items per page.
- W4385572734 abstract "Recent work has shown that Pre-trained Language Models (PLMs) store the relational knowledge learned from data and utilize it for performing downstream tasks. However, commonsense knowledge across different regions may vary. For instance, the color of bridal dress is white in American weddings whereas it is red in Chinese weddings. In this paper, we introduce a benchmark dataset, Geo-diverse Commonsense Multilingual Language Models Analysis (GeoMLAMA), for probing the diversity of the relational knowledge in multilingual PLMs. GeoMLAMA contains 3125 prompts in English, Chinese, Hindi, Persian, and Swahili, with a wide coverage of concepts shared by people from American, Chinese, Indian, Iranian and Kenyan cultures. We benchmark 11 standard multilingual PLMs on GeoMLAMA. Interestingly, we find that 1) larger multilingual PLMs variants do not necessarily store geo-diverse concepts better than its smaller variant; 2) multilingual PLMs are not intrinsically biased towards knowledge from the Western countries (the United States); 3) the native language of a country may not be the best language to probe its knowledge and 4) a language may better probe knowledge about a non-native country than its native country." @default.
- W4385572734 created "2023-08-05" @default.
- W4385572734 creator A5004824034 @default.
- W4385572734 creator A5043693897 @default.
- W4385572734 creator A5048920247 @default.
- W4385572734 creator A5072376756 @default.
- W4385572734 creator A5087096372 @default.
- W4385572734 date "2022-01-01" @default.
- W4385572734 modified "2023-09-25" @default.
- W4385572734 title "GeoMLAMA: Geo-Diverse Commonsense Probing on Multilingual Pre-Trained Language Models" @default.
- W4385572734 doi "https://doi.org/10.18653/v1/2022.emnlp-main.132" @default.
- W4385572734 hasPublicationYear "2022" @default.
- W4385572734 type Work @default.
- W4385572734 citedByCount "1" @default.
- W4385572734 countsByYear W43855727342023 @default.
- W4385572734 crossrefType "proceedings-article" @default.
- W4385572734 hasAuthorship W4385572734A5004824034 @default.
- W4385572734 hasAuthorship W4385572734A5043693897 @default.
- W4385572734 hasAuthorship W4385572734A5048920247 @default.
- W4385572734 hasAuthorship W4385572734A5072376756 @default.
- W4385572734 hasAuthorship W4385572734A5087096372 @default.
- W4385572734 hasBestOaLocation W43855727341 @default.
- W4385572734 hasConcept C120567893 @default.
- W4385572734 hasConcept C13280743 @default.
- W4385572734 hasConcept C138885662 @default.
- W4385572734 hasConcept C154945302 @default.
- W4385572734 hasConcept C185798385 @default.
- W4385572734 hasConcept C204321447 @default.
- W4385572734 hasConcept C205649164 @default.
- W4385572734 hasConcept C2779913364 @default.
- W4385572734 hasConcept C30542707 @default.
- W4385572734 hasConcept C41008148 @default.
- W4385572734 hasConcept C41895202 @default.
- W4385572734 hasConcept C519982507 @default.
- W4385572734 hasConceptScore W4385572734C120567893 @default.
- W4385572734 hasConceptScore W4385572734C13280743 @default.
- W4385572734 hasConceptScore W4385572734C138885662 @default.
- W4385572734 hasConceptScore W4385572734C154945302 @default.
- W4385572734 hasConceptScore W4385572734C185798385 @default.
- W4385572734 hasConceptScore W4385572734C204321447 @default.
- W4385572734 hasConceptScore W4385572734C205649164 @default.
- W4385572734 hasConceptScore W4385572734C2779913364 @default.
- W4385572734 hasConceptScore W4385572734C30542707 @default.
- W4385572734 hasConceptScore W4385572734C41008148 @default.
- W4385572734 hasConceptScore W4385572734C41895202 @default.
- W4385572734 hasConceptScore W4385572734C519982507 @default.
- W4385572734 hasLocation W43855727341 @default.
- W4385572734 hasLocation W43855727342 @default.
- W4385572734 hasOpenAccess W4385572734 @default.
- W4385572734 hasPrimaryLocation W43855727341 @default.
- W4385572734 hasRelatedWork W112744582 @default.
- W4385572734 hasRelatedWork W1485630101 @default.
- W4385572734 hasRelatedWork W2081245617 @default.
- W4385572734 hasRelatedWork W2096720918 @default.
- W4385572734 hasRelatedWork W2498017833 @default.
- W4385572734 hasRelatedWork W2507895645 @default.
- W4385572734 hasRelatedWork W2789919619 @default.
- W4385572734 hasRelatedWork W3033750096 @default.
- W4385572734 hasRelatedWork W4385572734 @default.
- W4385572734 hasRelatedWork W4385611867 @default.
- W4385572734 isParatext "false" @default.
- W4385572734 isRetracted "false" @default.
- W4385572734 workType "article" @default.