Matches in SemOpenAlex for { <https://semopenalex.org/work/W3102684881> ?p ?o ?g. }
Showing items 1 to 88 of
88
with 100 items per page.
- W3102684881 abstract "Neural language models learn, to varying degrees of accuracy, the grammatical properties of natural languages. In this work, we investigate whether there are systematic sources of variation in the language models’ accuracy. Focusing on subject-verb agreement and reflexive anaphora, we find that certain nouns are systematically understood better than others, an effect which is robust across grammatical tasks and different language models. Surprisingly, we find that across four orders of magnitude, corpus frequency is unrelated to a noun’s performance on grammatical tasks. Finally, we find that a novel noun’s grammatical properties can be few-shot learned from various types of training data. The results present a paradox: there should be less variation in grammatical performance than is actually observed." @default.
- W3102684881 created "2020-11-23" @default.
- W3102684881 creator A5025743443 @default.
- W3102684881 creator A5039944628 @default.
- W3102684881 creator A5072177784 @default.
- W3102684881 creator A5078977334 @default.
- W3102684881 date "2020-01-01" @default.
- W3102684881 modified "2023-10-02" @default.
- W3102684881 title "Word Frequency Does Not Predict Grammatical Knowledge in Language Models" @default.
- W3102684881 cites W2036963181 @default.
- W3102684881 cites W2038721957 @default.
- W3102684881 cites W2039217078 @default.
- W3102684881 cites W2060277733 @default.
- W3102684881 cites W2193149 @default.
- W3102684881 cites W2340992439 @default.
- W3102684881 cites W2531882892 @default.
- W3102684881 cites W2549835527 @default.
- W3102684881 cites W2798727047 @default.
- W3102684881 cites W2888922637 @default.
- W3102684881 cites W2910243263 @default.
- W3102684881 cites W2918996109 @default.
- W3102684881 cites W2921890305 @default.
- W3102684881 cites W2962961857 @default.
- W3102684881 cites W2963026768 @default.
- W3102684881 cites W2963341956 @default.
- W3102684881 cites W2963751529 @default.
- W3102684881 cites W2964110616 @default.
- W3102684881 cites W2964222268 @default.
- W3102684881 cites W2970597249 @default.
- W3102684881 cites W2978670439 @default.
- W3102684881 cites W3035295386 @default.
- W3102684881 cites W75403801 @default.
- W3102684881 doi "https://doi.org/10.18653/v1/2020.emnlp-main.331" @default.
- W3102684881 hasPublicationYear "2020" @default.
- W3102684881 type Work @default.
- W3102684881 sameAs 3102684881 @default.
- W3102684881 citedByCount "6" @default.
- W3102684881 countsByYear W31026848812021 @default.
- W3102684881 countsByYear W31026848812022 @default.
- W3102684881 countsByYear W31026848812023 @default.
- W3102684881 crossrefType "proceedings-article" @default.
- W3102684881 hasAuthorship W3102684881A5025743443 @default.
- W3102684881 hasAuthorship W3102684881A5039944628 @default.
- W3102684881 hasAuthorship W3102684881A5072177784 @default.
- W3102684881 hasAuthorship W3102684881A5078977334 @default.
- W3102684881 hasBestOaLocation W31026848811 @default.
- W3102684881 hasConcept C121332964 @default.
- W3102684881 hasConcept C121934690 @default.
- W3102684881 hasConcept C138885662 @default.
- W3102684881 hasConcept C154945302 @default.
- W3102684881 hasConcept C204321447 @default.
- W3102684881 hasConcept C206977991 @default.
- W3102684881 hasConcept C2776397901 @default.
- W3102684881 hasConcept C2778334786 @default.
- W3102684881 hasConcept C41008148 @default.
- W3102684881 hasConcept C41895202 @default.
- W3102684881 hasConcept C44870925 @default.
- W3102684881 hasConcept C90805587 @default.
- W3102684881 hasConceptScore W3102684881C121332964 @default.
- W3102684881 hasConceptScore W3102684881C121934690 @default.
- W3102684881 hasConceptScore W3102684881C138885662 @default.
- W3102684881 hasConceptScore W3102684881C154945302 @default.
- W3102684881 hasConceptScore W3102684881C204321447 @default.
- W3102684881 hasConceptScore W3102684881C206977991 @default.
- W3102684881 hasConceptScore W3102684881C2776397901 @default.
- W3102684881 hasConceptScore W3102684881C2778334786 @default.
- W3102684881 hasConceptScore W3102684881C41008148 @default.
- W3102684881 hasConceptScore W3102684881C41895202 @default.
- W3102684881 hasConceptScore W3102684881C44870925 @default.
- W3102684881 hasConceptScore W3102684881C90805587 @default.
- W3102684881 hasLocation W31026848811 @default.
- W3102684881 hasLocation W31026848812 @default.
- W3102684881 hasOpenAccess W3102684881 @default.
- W3102684881 hasPrimaryLocation W31026848811 @default.
- W3102684881 hasRelatedWork W1538709803 @default.
- W3102684881 hasRelatedWork W1968323823 @default.
- W3102684881 hasRelatedWork W2017675264 @default.
- W3102684881 hasRelatedWork W2373756576 @default.
- W3102684881 hasRelatedWork W2602782736 @default.
- W3102684881 hasRelatedWork W2796206592 @default.
- W3102684881 hasRelatedWork W2990733512 @default.
- W3102684881 hasRelatedWork W3023303032 @default.
- W3102684881 hasRelatedWork W3154636470 @default.
- W3102684881 hasRelatedWork W4385999727 @default.
- W3102684881 isParatext "false" @default.
- W3102684881 isRetracted "false" @default.
- W3102684881 magId "3102684881" @default.
- W3102684881 workType "article" @default.