Matches in SemOpenAlex for { <https://semopenalex.org/work/W2991225581> ?p ?o ?g. }
Showing items 1 to 93 of
93
with 100 items per page.
- W2991225581 abstract "Recently, data-driven task-oriented dialogue systems have achieved promising performance in English. However, developing dialogue systems that support low-resource languages remains a long-standing challenge due to the absence of high-quality data. In order to circumvent the expensive and time-consuming data collection, we introduce Attention-Informed Mixed-Language Training (MLT), a novel zero-shot adaptation method for cross-lingual task-oriented dialogue systems. It leverages very few task-related parallel word pairs to generate code-switching sentences for learning the inter-lingual semantics across languages. Instead of manually selecting the word pairs, we propose to extract source words based on the scores computed by the attention layer of a trained English task-related model and then generate word pairs using existing bilingual dictionaries. Furthermore, intensive experiments with different cross-lingual embeddings demonstrate the effectiveness of our approach. Finally, with very few word pairs, our model achieves significant zero-shot adaptation performance improvements in both cross-lingual dialogue state tracking and natural language understanding (i.e., intent detection and slot filling) tasks compared to the current state-of-the-art approaches, which utilize a much larger amount of bilingual data." @default.
- W2991225581 created "2019-12-05" @default.
- W2991225581 creator A5011760791 @default.
- W2991225581 creator A5021691940 @default.
- W2991225581 creator A5065856469 @default.
- W2991225581 creator A5077532535 @default.
- W2991225581 creator A5085516032 @default.
- W2991225581 date "2019-11-20" @default.
- W2991225581 modified "2023-10-01" @default.
- W2991225581 title "Attention-Informed Mixed-Language Training for Zero-shot Cross-lingual Task-oriented Dialogue Systems" @default.
- W2991225581 cites W2296283641 @default.
- W2991225581 cites W2471692228 @default.
- W2991225581 cites W2493916176 @default.
- W2991225581 cites W2620558438 @default.
- W2991225581 cites W2734693922 @default.
- W2991225581 cites W2742113707 @default.
- W2991225581 cites W2758640791 @default.
- W2991225581 cites W2791169651 @default.
- W2991225581 cites W2798367796 @default.
- W2991225581 cites W2803392141 @default.
- W2991225581 cites W2866343820 @default.
- W2991225581 cites W2885323099 @default.
- W2991225581 cites W2891896107 @default.
- W2991225581 cites W2898856000 @default.
- W2991225581 cites W2914120296 @default.
- W2991225581 cites W2950765829 @default.
- W2991225581 cites W2962738716 @default.
- W2991225581 cites W2963118869 @default.
- W2991225581 cites W2963341956 @default.
- W2991225581 cites W2963974889 @default.
- W2991225581 cites W2964101860 @default.
- W2991225581 cites W2971160427 @default.
- W2991225581 cites W3106003309 @default.
- W2991225581 doi "https://doi.org/10.48550/arxiv.1911.09273" @default.
- W2991225581 hasPublicationYear "2019" @default.
- W2991225581 type Work @default.
- W2991225581 sameAs 2991225581 @default.
- W2991225581 citedByCount "7" @default.
- W2991225581 countsByYear W29912255812020 @default.
- W2991225581 crossrefType "posted-content" @default.
- W2991225581 hasAuthorship W2991225581A5011760791 @default.
- W2991225581 hasAuthorship W2991225581A5021691940 @default.
- W2991225581 hasAuthorship W2991225581A5065856469 @default.
- W2991225581 hasAuthorship W2991225581A5077532535 @default.
- W2991225581 hasAuthorship W2991225581A5085516032 @default.
- W2991225581 hasBestOaLocation W29912255811 @default.
- W2991225581 hasConcept C120665830 @default.
- W2991225581 hasConcept C121332964 @default.
- W2991225581 hasConcept C138885662 @default.
- W2991225581 hasConcept C139807058 @default.
- W2991225581 hasConcept C154945302 @default.
- W2991225581 hasConcept C162324750 @default.
- W2991225581 hasConcept C184337299 @default.
- W2991225581 hasConcept C187736073 @default.
- W2991225581 hasConcept C199360897 @default.
- W2991225581 hasConcept C204321447 @default.
- W2991225581 hasConcept C2780451532 @default.
- W2991225581 hasConcept C2780813799 @default.
- W2991225581 hasConcept C41008148 @default.
- W2991225581 hasConcept C41895202 @default.
- W2991225581 hasConcept C90805587 @default.
- W2991225581 hasConceptScore W2991225581C120665830 @default.
- W2991225581 hasConceptScore W2991225581C121332964 @default.
- W2991225581 hasConceptScore W2991225581C138885662 @default.
- W2991225581 hasConceptScore W2991225581C139807058 @default.
- W2991225581 hasConceptScore W2991225581C154945302 @default.
- W2991225581 hasConceptScore W2991225581C162324750 @default.
- W2991225581 hasConceptScore W2991225581C184337299 @default.
- W2991225581 hasConceptScore W2991225581C187736073 @default.
- W2991225581 hasConceptScore W2991225581C199360897 @default.
- W2991225581 hasConceptScore W2991225581C204321447 @default.
- W2991225581 hasConceptScore W2991225581C2780451532 @default.
- W2991225581 hasConceptScore W2991225581C2780813799 @default.
- W2991225581 hasConceptScore W2991225581C41008148 @default.
- W2991225581 hasConceptScore W2991225581C41895202 @default.
- W2991225581 hasConceptScore W2991225581C90805587 @default.
- W2991225581 hasLocation W29912255811 @default.
- W2991225581 hasOpenAccess W2991225581 @default.
- W2991225581 hasPrimaryLocation W29912255811 @default.
- W2991225581 hasRelatedWork W1539050421 @default.
- W2991225581 hasRelatedWork W1563147278 @default.
- W2991225581 hasRelatedWork W1571518467 @default.
- W2991225581 hasRelatedWork W2020291234 @default.
- W2991225581 hasRelatedWork W2081647779 @default.
- W2991225581 hasRelatedWork W2082609543 @default.
- W2991225581 hasRelatedWork W2360025963 @default.
- W2991225581 hasRelatedWork W2360785147 @default.
- W2991225581 hasRelatedWork W2962901450 @default.
- W2991225581 hasRelatedWork W87991986 @default.
- W2991225581 isParatext "false" @default.
- W2991225581 isRetracted "false" @default.
- W2991225581 magId "2991225581" @default.
- W2991225581 workType "article" @default.