Matches in SemOpenAlex for { <https://semopenalex.org/work/W3092162010> ?p ?o ?g. }
- W3092162010 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.
- W3092162010 created "2020-10-15" @default.
- W3092162010 creator A5049259877 @default.
- W3092162010 creator A5061107628 @default.
- W3092162010 creator A5070625077 @default.
- W3092162010 creator A5088803982 @default.
- W3092162010 date "2020-10-07" @default.
- W3092162010 modified "2023-09-23" @default.
- W3092162010 title "Improving Context Modeling in Neural Topic Segmentation" @default.
- W3092162010 cites W124349300 @default.
- W3092162010 cites W1557074680 @default.
- W3092162010 cites W1598796236 @default.
- W3092162010 cites W160817811 @default.
- W3092162010 cites W1626945812 @default.
- W3092162010 cites W1781198478 @default.
- W3092162010 cites W1828401780 @default.
- W3092162010 cites W1862888253 @default.
- W3092162010 cites W2017776283 @default.
- W3092162010 cites W2027823133 @default.
- W3092162010 cites W2106918957 @default.
- W3092162010 cites W2118612506 @default.
- W3092162010 cites W2131774270 @default.
- W3092162010 cites W2140676672 @default.
- W3092162010 cites W2164439758 @default.
- W3092162010 cites W2470673105 @default.
- W3092162010 cites W2512217112 @default.
- W3092162010 cites W2512597464 @default.
- W3092162010 cites W2559122826 @default.
- W3092162010 cites W2758753219 @default.
- W3092162010 cites W2759200442 @default.
- W3092162010 cites W2761764495 @default.
- W3092162010 cites W2793710388 @default.
- W3092162010 cites W2808114373 @default.
- W3092162010 cites W2889446948 @default.
- W3092162010 cites W2914694065 @default.
- W3092162010 cites W2921642681 @default.
- W3092162010 cites W2949573100 @default.
- W3092162010 cites W2962716111 @default.
- W3092162010 cites W2963341956 @default.
- W3092162010 cites W2963403868 @default.
- W3092162010 cites W2966697108 @default.
- W3092162010 cites W2970008578 @default.
- W3092162010 cites W2970263339 @default.
- W3092162010 cites W2997805114 @default.
- W3092162010 cites W3034327408 @default.
- W3092162010 cites W3034594820 @default.
- W3092162010 cites W3099609223 @default.
- W3092162010 cites W3102373121 @default.
- W3092162010 cites W3123587674 @default.
- W3092162010 cites W3176113532 @default.
- W3092162010 hasPublicationYear "2020" @default.
- W3092162010 type Work @default.
- W3092162010 sameAs 3092162010 @default.
- W3092162010 citedByCount "0" @default.
- W3092162010 crossrefType "posted-content" @default.
- W3092162010 hasAuthorship W3092162010A5049259877 @default.
- W3092162010 hasAuthorship W3092162010A5061107628 @default.
- W3092162010 hasAuthorship W3092162010A5070625077 @default.
- W3092162010 hasAuthorship W3092162010A5088803982 @default.
- W3092162010 hasConcept C104317684 @default.
- W3092162010 hasConcept C119857082 @default.
- W3092162010 hasConcept C150899416 @default.
- W3092162010 hasConcept C151730666 @default.
- W3092162010 hasConcept C154775046 @default.
- W3092162010 hasConcept C154945302 @default.
- W3092162010 hasConcept C162324750 @default.
- W3092162010 hasConcept C166957645 @default.
- W3092162010 hasConcept C185592680 @default.
- W3092162010 hasConcept C187736073 @default.
- W3092162010 hasConcept C204321447 @default.
- W3092162010 hasConcept C26517878 @default.
- W3092162010 hasConcept C2779343474 @default.
- W3092162010 hasConcept C2780451532 @default.
- W3092162010 hasConcept C2984842247 @default.
- W3092162010 hasConcept C38652104 @default.
- W3092162010 hasConcept C41008148 @default.
- W3092162010 hasConcept C50644808 @default.
- W3092162010 hasConcept C55493867 @default.
- W3092162010 hasConcept C63479239 @default.
- W3092162010 hasConcept C86803240 @default.
- W3092162010 hasConcept C89600930 @default.
- W3092162010 hasConcept C95457728 @default.
- W3092162010 hasConceptScore W3092162010C104317684 @default.
- W3092162010 hasConceptScore W3092162010C119857082 @default.
- W3092162010 hasConceptScore W3092162010C150899416 @default.
- W3092162010 hasConceptScore W3092162010C151730666 @default.
- W3092162010 hasConceptScore W3092162010C154775046 @default.
- W3092162010 hasConceptScore W3092162010C154945302 @default.
- W3092162010 hasConceptScore W3092162010C162324750 @default.
- W3092162010 hasConceptScore W3092162010C166957645 @default.
- W3092162010 hasConceptScore W3092162010C185592680 @default.
- W3092162010 hasConceptScore W3092162010C187736073 @default.
- W3092162010 hasConceptScore W3092162010C204321447 @default.
- W3092162010 hasConceptScore W3092162010C26517878 @default.
- W3092162010 hasConceptScore W3092162010C2779343474 @default.
- W3092162010 hasConceptScore W3092162010C2780451532 @default.
- W3092162010 hasConceptScore W3092162010C2984842247 @default.
- W3092162010 hasConceptScore W3092162010C38652104 @default.
- W3092162010 hasConceptScore W3092162010C41008148 @default.
- W3092162010 hasConceptScore W3092162010C50644808 @default.