FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3103427938> ?p ?o ?g. }

• W3103427938 abstract "We analyze several recent unsupervised constituency parsing models, which are tuned with respect to the parsing F1 score on the Wall Street Journal (WSJ) development set (1,700 sentences). We introduce strong baselines for them, by training an existing supervised parsing model (Kitaev and Klein, 2018) on the same labeled examples they access. When training on the 1,700 examples, or even when using only 50 examples for training and 5 for development, such a few-shot parsing approach can outperform all the unsupervised parsing methods by a significant margin. Few-shot parsing can be further improved by a simple data augmentation method and self-training. This suggests that, in order to arrive at fair conclusions, we should carefully consider the amount of labeled data used for model development. We propose two protocols for future work on unsupervised parsing: (i) use fully unsupervised criteria for hyperparameter tuning and model selection; (ii) use as few labeled examples as possible for model development, and compare to few-shot parsing trained on the same labeled examples." @default.
• W3103427938 created "2020-11-23" @default.
• W3103427938 creator A5015602781 @default.
• W3103427938 creator A5081022650 @default.
• W3103427938 creator A5091670413 @default.
• W3103427938 date "2020-01-01" @default.
• W3103427938 modified "2023-09-26" @default.
• W3103427938 title "On the Role of Supervision in Unsupervised Constituency Parsing" @default.
• W3103427938 cites W1527783480 @default.
• W3103427938 cites W1632114991 @default.
• W3103427938 cites W176283576 @default.
• W3103427938 cites W1848000237 @default.
• W3103427938 cites W1879966306 @default.
• W3103427938 cites W1967147081 @default.
• W3103427938 cites W2002586403 @default.
• W3103427938 cites W2005085483 @default.
• W3103427938 cites W2064675550 @default.
• W3103427938 cites W2102366626 @default.
• W3103427938 cites W2115803233 @default.
• W3103427938 cites W2129882630 @default.
• W3103427938 cites W2137807494 @default.
• W3103427938 cites W2137807925 @default.
• W3103427938 cites W2145837098 @default.
• W3103427938 cites W2157762871 @default.
• W3103427938 cites W2161839092 @default.
• W3103427938 cites W2163568299 @default.
• W3103427938 cites W2175969983 @default.
• W3103427938 cites W2250520759 @default.
• W3103427938 cites W2252238553 @default.
• W3103427938 cites W2798569372 @default.
• W3103427938 cites W2817457237 @default.
• W3103427938 cites W28418033 @default.
• W3103427938 cites W2889260178 @default.
• W3103427938 cites W2899771611 @default.
• W3103427938 cites W2932376173 @default.
• W3103427938 cites W2932637973 @default.
• W3103427938 cites W2949111363 @default.
• W3103427938 cites W2949399644 @default.
• W3103427938 cites W2949579048 @default.
• W3103427938 cites W2950896474 @default.
• W3103427938 cites W2962739339 @default.
• W3103427938 cites W2963073938 @default.
• W3103427938 cites W2963084773 @default.
• W3103427938 cites W2963355447 @default.
• W3103427938 cites W2963403868 @default.
• W3103427938 cites W2963411763 @default.
• W3103427938 cites W2963522904 @default.
• W3103427938 cites W2963525616 @default.
• W3103427938 cites W2963545917 @default.
• W3103427938 cites W2963754491 @default.
• W3103427938 cites W2963863610 @default.
• W3103427938 cites W2964087600 @default.
• W3103427938 cites W2964121744 @default.
• W3103427938 cites W2970365965 @default.
• W3103427938 cites W2970378492 @default.
• W3103427938 cites W2971351900 @default.
• W3103427938 cites W2996493817 @default.
• W3103427938 cites W3016560063 @default.
• W3103427938 doi "https://doi.org/10.18653/v1/2020.emnlp-main.614" @default.
• W3103427938 hasPublicationYear "2020" @default.
• W3103427938 type Work @default.
• W3103427938 sameAs 3103427938 @default.
• W3103427938 citedByCount "9" @default.
• W3103427938 countsByYear W31034279382020 @default.
• W3103427938 countsByYear W31034279382021 @default.
• W3103427938 countsByYear W31034279382022 @default.
• W3103427938 crossrefType "proceedings-article" @default.
• W3103427938 hasAuthorship W3103427938A5015602781 @default.
• W3103427938 hasAuthorship W3103427938A5081022650 @default.
• W3103427938 hasAuthorship W3103427938A5091670413 @default.
• W3103427938 hasBestOaLocation W31034279381 @default.
• W3103427938 hasConcept C118364021 @default.
• W3103427938 hasConcept C119857082 @default.
• W3103427938 hasConcept C154945302 @default.
• W3103427938 hasConcept C177264268 @default.
• W3103427938 hasConcept C186644900 @default.
• W3103427938 hasConcept C199360897 @default.
• W3103427938 hasConcept C204321447 @default.
• W3103427938 hasConcept C41008148 @default.
• W3103427938 hasConcept C42560504 @default.
• W3103427938 hasConcept C60690694 @default.
• W3103427938 hasConcept C774472 @default.
• W3103427938 hasConcept C8038995 @default.
• W3103427938 hasConcept C8642999 @default.
• W3103427938 hasConceptScore W3103427938C118364021 @default.
• W3103427938 hasConceptScore W3103427938C119857082 @default.
• W3103427938 hasConceptScore W3103427938C154945302 @default.
• W3103427938 hasConceptScore W3103427938C177264268 @default.
• W3103427938 hasConceptScore W3103427938C186644900 @default.
• W3103427938 hasConceptScore W3103427938C199360897 @default.
• W3103427938 hasConceptScore W3103427938C204321447 @default.
• W3103427938 hasConceptScore W3103427938C41008148 @default.
• W3103427938 hasConceptScore W3103427938C42560504 @default.
• W3103427938 hasConceptScore W3103427938C60690694 @default.
• W3103427938 hasConceptScore W3103427938C774472 @default.
• W3103427938 hasConceptScore W3103427938C8038995 @default.
• W3103427938 hasConceptScore W3103427938C8642999 @default.
• W3103427938 hasLocation W31034279381 @default.
• W3103427938 hasLocation W31034279382 @default.
• W3103427938 hasOpenAccess W3103427938 @default.

• next