Matches in SemOpenAlex for { <https://semopenalex.org/work/W3115355887> ?p ?o ?g. }
- W3115355887 endingPage "636" @default.
- W3115355887 startingPage "626" @default.
- W3115355887 abstract "Topic segmentation is critical in key NLP tasks and recent works favor highly effective neural supervised approaches. However, current neural solutions are arguably limited in how they model context. In this paper, we enhance a segmenter based on a hierarchical attention BiLSTM network to better model context, by adding a coherence-related auxiliary task and restricted self-attention. Our optimized segmenter outperforms SOTA approaches when trained and tested on three datasets. We also the robustness of our proposed model in domain transfer setting by training a model on a large-scale dataset and testing it on four challenging real-world benchmarks. Furthermore, we apply our proposed strategy to two other languages (German and Chinese), and show its effectiveness in multilingual scenarios." @default.
- W3115355887 created "2021-01-05" @default.
- W3115355887 creator A5049259877 @default.
- W3115355887 creator A5061107628 @default.
- W3115355887 creator A5070625077 @default.
- W3115355887 creator A5088803982 @default.
- W3115355887 date "2020-12-01" @default.
- W3115355887 modified "2023-09-23" @default.
- W3115355887 title "Improving Context Modeling in Neural Topic Segmentation." @default.
- W3115355887 cites W124349300 @default.
- W3115355887 cites W1557074680 @default.
- W3115355887 cites W1598796236 @default.
- W3115355887 cites W160817811 @default.
- W3115355887 cites W1626945812 @default.
- W3115355887 cites W1781198478 @default.
- W3115355887 cites W1828401780 @default.
- W3115355887 cites W1862888253 @default.
- W3115355887 cites W2017776283 @default.
- W3115355887 cites W2027823133 @default.
- W3115355887 cites W2106918957 @default.
- W3115355887 cites W2118612506 @default.
- W3115355887 cites W2131774270 @default.
- W3115355887 cites W2140676672 @default.
- W3115355887 cites W2164439758 @default.
- W3115355887 cites W2470673105 @default.
- W3115355887 cites W2512217112 @default.
- W3115355887 cites W2512597464 @default.
- W3115355887 cites W2559122826 @default.
- W3115355887 cites W2758753219 @default.
- W3115355887 cites W2759200442 @default.
- W3115355887 cites W2761764495 @default.
- W3115355887 cites W2793710388 @default.
- W3115355887 cites W2808114373 @default.
- W3115355887 cites W2889446948 @default.
- W3115355887 cites W2914694065 @default.
- W3115355887 cites W2921642681 @default.
- W3115355887 cites W2949573100 @default.
- W3115355887 cites W2962716111 @default.
- W3115355887 cites W2963341956 @default.
- W3115355887 cites W2963403868 @default.
- W3115355887 cites W2970008578 @default.
- W3115355887 cites W2970263339 @default.
- W3115355887 cites W2997805114 @default.
- W3115355887 cites W3034327408 @default.
- W3115355887 cites W3034594820 @default.
- W3115355887 cites W3099609223 @default.
- W3115355887 cites W3102373121 @default.
- W3115355887 cites W3123587674 @default.
- W3115355887 cites W3176113532 @default.
- W3115355887 hasPublicationYear "2020" @default.
- W3115355887 type Work @default.
- W3115355887 sameAs 3115355887 @default.
- W3115355887 citedByCount "5" @default.
- W3115355887 countsByYear W31153558872021 @default.
- W3115355887 crossrefType "proceedings-article" @default.
- W3115355887 hasAuthorship W3115355887A5049259877 @default.
- W3115355887 hasAuthorship W3115355887A5061107628 @default.
- W3115355887 hasAuthorship W3115355887A5070625077 @default.
- W3115355887 hasAuthorship W3115355887A5088803982 @default.
- W3115355887 hasConcept C104317684 @default.
- W3115355887 hasConcept C119857082 @default.
- W3115355887 hasConcept C150899416 @default.
- W3115355887 hasConcept C151730666 @default.
- W3115355887 hasConcept C154945302 @default.
- W3115355887 hasConcept C162324750 @default.
- W3115355887 hasConcept C183322885 @default.
- W3115355887 hasConcept C185592680 @default.
- W3115355887 hasConcept C187736073 @default.
- W3115355887 hasConcept C204321447 @default.
- W3115355887 hasConcept C26517878 @default.
- W3115355887 hasConcept C2779343474 @default.
- W3115355887 hasConcept C2780451532 @default.
- W3115355887 hasConcept C2781238097 @default.
- W3115355887 hasConcept C38652104 @default.
- W3115355887 hasConcept C41008148 @default.
- W3115355887 hasConcept C50644808 @default.
- W3115355887 hasConcept C55493867 @default.
- W3115355887 hasConcept C63479239 @default.
- W3115355887 hasConcept C86803240 @default.
- W3115355887 hasConcept C89600930 @default.
- W3115355887 hasConceptScore W3115355887C104317684 @default.
- W3115355887 hasConceptScore W3115355887C119857082 @default.
- W3115355887 hasConceptScore W3115355887C150899416 @default.
- W3115355887 hasConceptScore W3115355887C151730666 @default.
- W3115355887 hasConceptScore W3115355887C154945302 @default.
- W3115355887 hasConceptScore W3115355887C162324750 @default.
- W3115355887 hasConceptScore W3115355887C183322885 @default.
- W3115355887 hasConceptScore W3115355887C185592680 @default.
- W3115355887 hasConceptScore W3115355887C187736073 @default.
- W3115355887 hasConceptScore W3115355887C204321447 @default.
- W3115355887 hasConceptScore W3115355887C26517878 @default.
- W3115355887 hasConceptScore W3115355887C2779343474 @default.
- W3115355887 hasConceptScore W3115355887C2780451532 @default.
- W3115355887 hasConceptScore W3115355887C2781238097 @default.
- W3115355887 hasConceptScore W3115355887C38652104 @default.
- W3115355887 hasConceptScore W3115355887C41008148 @default.
- W3115355887 hasConceptScore W3115355887C50644808 @default.
- W3115355887 hasConceptScore W3115355887C55493867 @default.