FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3213683640> ?p ?o ?g. }

• W3213683640 endingPage "1670" @default.
• W3213683640 startingPage "1655" @default.
• W3213683640 abstract "We present a simple but effective approach for leveraging Wikipedia for neural machine translation as well as cross-lingual tasks of image captioning and dependency parsing without using any direct supervision from external parallel data or supervised models in the target language. We show that first sentences and titles of linked Wikipedia pages, as well as cross-lingual image captions, are strong signals for a seed parallel data to extract bilingual dictionaries and cross-lingual word embeddings for mining parallel text from Wikipedia. Our final model achieves high BLEU scores that are close to or sometimes higher than strong supervised baselines in low-resource languages; e.g. supervised BLEU of 4.0 versus 12.1 from our model in English-to-Kazakh. Moreover, we tailor our wikily translation models to unsupervised image captioning, and cross-lingual dependency parser transfer. In image captioning, we train a multi-tasking machine translation and image captioning pipeline for Arabic and English from which the Arabic training data is a wikily translation of the English captioning data. Our captioning results on Arabic are slightly better than that of its supervised model. In dependency parsing, we translate a large amount of monolingual text, and use it as an artificial training data in an annotation projection framework. We show that our model outperforms recent work on cross-lingual transfer of dependency parsers." @default.
• W3213683640 created "2021-11-22" @default.
• W3213683640 creator A5027174211 @default.
• W3213683640 creator A5068508539 @default.
• W3213683640 creator A5087104756 @default.
• W3213683640 date "2021-11-01" @default.
• W3213683640 modified "2023-09-26" @default.
• W3213683640 title "“Wikily” Supervised Neural Translation Tailored to Cross-Lingual Tasks" @default.
• W3213683640 cites W1249771036 @default.
• W3213683640 cites W1574126082 @default.
• W3213683640 cites W1636405317 @default.
• W3213683640 cites W1726342440 @default.
• W3213683640 cites W1889081078 @default.
• W3213683640 cites W2016630033 @default.
• W3213683640 cites W2047295649 @default.
• W3213683640 cites W2101105183 @default.
• W3213683640 cites W2114609248 @default.
• W3213683640 cites W2114879013 @default.
• W3213683640 cites W2124807415 @default.
• W3213683640 cites W2143954309 @default.
• W3213683640 cites W2148708890 @default.
• W3213683640 cites W2156985047 @default.
• W3213683640 cites W2157331557 @default.
• W3213683640 cites W2194775991 @default.
• W3213683640 cites W22168010 @default.
• W3213683640 cites W2250313959 @default.
• W3213683640 cites W2250681176 @default.
• W3213683640 cites W2509282593 @default.
• W3213683640 cites W2512924740 @default.
• W3213683640 cites W2538197161 @default.
• W3213683640 cites W2572474373 @default.
• W3213683640 cites W2574640638 @default.
• W3213683640 cites W2757521750 @default.
• W3213683640 cites W2782590789 @default.
• W3213683640 cites W2884476676 @default.
• W3213683640 cites W2886198413 @default.
• W3213683640 cites W2886641317 @default.
• W3213683640 cites W2889326796 @default.
• W3213683640 cites W2890007195 @default.
• W3213683640 cites W2914096745 @default.
• W3213683640 cites W2922197727 @default.
• W3213683640 cites W2944815030 @default.
• W3213683640 cites W2949405462 @default.
• W3213683640 cites W2949911645 @default.
• W3213683640 cites W2950485982 @default.
• W3213683640 cites W2952768586 @default.
• W3213683640 cites W2960374072 @default.
• W3213683640 cites W2962735107 @default.
• W3213683640 cites W2962801832 @default.
• W3213683640 cites W2962824887 @default.
• W3213683640 cites W2963048642 @default.
• W3213683640 cites W2963088995 @default.
• W3213683640 cites W2963216553 @default.
• W3213683640 cites W2963250244 @default.
• W3213683640 cites W2963341956 @default.
• W3213683640 cites W2963360627 @default.
• W3213683640 cites W2963403868 @default.
• W3213683640 cites W2963532001 @default.
• W3213683640 cites W2963602293 @default.
• W3213683640 cites W2963633299 @default.
• W3213683640 cites W2963667932 @default.
• W3213683640 cites W2963743213 @default.
• W3213683640 cites W2963877297 @default.
• W3213683640 cites W2964073484 @default.
• W3213683640 cites W2964121744 @default.
• W3213683640 cites W2970971581 @default.
• W3213683640 cites W2977458338 @default.
• W3213683640 cites W2981994675 @default.
• W3213683640 cites W2988451549 @default.
• W3213683640 cites W2992478697 @default.
• W3213683640 cites W2995460523 @default.
• W3213683640 cites W3011608779 @default.
• W3213683640 cites W3015354748 @default.
• W3213683640 cites W3016672431 @default.
• W3213683640 cites W3034773362 @default.
• W3213683640 cites W3037109418 @default.
• W3213683640 cites W3038734866 @default.
• W3213683640 cites W3082017874 @default.
• W3213683640 cites W3089109144 @default.
• W3213683640 cites W3089515120 @default.
• W3213683640 cites W3105378761 @default.
• W3213683640 cites W3107826490 @default.
• W3213683640 cites W3119378114 @default.
• W3213683640 cites W3122659018 @default.
• W3213683640 cites W3135099046 @default.
• W3213683640 cites W3135321472 @default.
• W3213683640 cites W3137065374 @default.
• W3213683640 cites W3167047789 @default.
• W3213683640 cites W3182683290 @default.
• W3213683640 cites W342285082 @default.
• W3213683640 cites W68733909 @default.
• W3213683640 hasPublicationYear "2021" @default.
• W3213683640 type Work @default.
• W3213683640 sameAs 3213683640 @default.
• W3213683640 citedByCount "0" @default.
• W3213683640 crossrefType "proceedings-article" @default.
• W3213683640 hasAuthorship W3213683640A5027174211 @default.
• W3213683640 hasAuthorship W3213683640A5068508539 @default.

• next