Matches in SemOpenAlex for { <https://semopenalex.org/work/W4281699648> ?p ?o ?g. }
Showing items 1 to 84 of
84
with 100 items per page.
- W4281699648 endingPage "542" @default.
- W4281699648 startingPage "521" @default.
- W4281699648 abstract "The aim of the study is to obtain a more lightweight language model that is comparable in terms of EM and F1 with the best modern language models in the task of finding the answer to a question in a text in Russian. The results of the work can be used in various question-and-answer systems for which response time is important. Since the lighter model has fewer parameters than the original one, it can be used on less powerful computing devices, including mobile devices. In this paper, methods of natural language processing, machine learning, and the theory of artificial neural networks are used. The neural network is configured and trained using the Torch and Hugging face machine learning libraries. In the work, the DistilBERT model was trained on the SberQUAD dataset with and without distillation. The work of the received models is compared. The distilled DistilBERT model (EM 58,57 and F1 78,42) was able to outperform the results of the larger ruGPT-3-medium generative network (EM 57,60 and F1 77,73), despite the fact that ruGPT-3-medium had 6,5 times more parameters. The model also showed better EM and F1 metrics than the same model, but to which only conventional training without distillation was applied (EM 55,65, F1 76,51). Unfortunately, the resulting model lags further behind the larger robert discriminative model (EM 66,83, F1 84,95), which has 3,2 times more parameters. The application of the DistilBERT model in question-and-answer systems in Russian is substantiated. Directions for further research are proposed." @default.
- W4281699648 created "2022-06-13" @default.
- W4281699648 creator A5061718419 @default.
- W4281699648 creator A5066804631 @default.
- W4281699648 date "2022-05-06" @default.
- W4281699648 modified "2023-09-26" @default.
- W4281699648 title "Experimental Study of Language Models of Transformer in the Problem of Finding the Answer to a Question in a Russian-Language Text" @default.
- W4281699648 cites W2121879602 @default.
- W4281699648 cites W2962784628 @default.
- W4281699648 cites W2963748441 @default.
- W4281699648 cites W2964110616 @default.
- W4281699648 cites W2983128379 @default.
- W4281699648 cites W3034999214 @default.
- W4281699648 cites W3035207248 @default.
- W4281699648 cites W3088049945 @default.
- W4281699648 cites W3100439847 @default.
- W4281699648 cites W3105234097 @default.
- W4281699648 cites W3117122667 @default.
- W4281699648 cites W3171778302 @default.
- W4281699648 cites W3175000455 @default.
- W4281699648 doi "https://doi.org/10.15622/ia.21.3.3" @default.
- W4281699648 hasPublicationYear "2022" @default.
- W4281699648 type Work @default.
- W4281699648 citedByCount "0" @default.
- W4281699648 crossrefType "journal-article" @default.
- W4281699648 hasAuthorship W4281699648A5061718419 @default.
- W4281699648 hasAuthorship W4281699648A5066804631 @default.
- W4281699648 hasBestOaLocation W42816996481 @default.
- W4281699648 hasConcept C119599485 @default.
- W4281699648 hasConcept C119857082 @default.
- W4281699648 hasConcept C127413603 @default.
- W4281699648 hasConcept C137293760 @default.
- W4281699648 hasConcept C154945302 @default.
- W4281699648 hasConcept C165801399 @default.
- W4281699648 hasConcept C195324797 @default.
- W4281699648 hasConcept C201995342 @default.
- W4281699648 hasConcept C203005215 @default.
- W4281699648 hasConcept C204321447 @default.
- W4281699648 hasConcept C2780451532 @default.
- W4281699648 hasConcept C39890363 @default.
- W4281699648 hasConcept C41008148 @default.
- W4281699648 hasConcept C44291984 @default.
- W4281699648 hasConcept C50644808 @default.
- W4281699648 hasConcept C622187 @default.
- W4281699648 hasConcept C66322947 @default.
- W4281699648 hasConcept C97931131 @default.
- W4281699648 hasConceptScore W4281699648C119599485 @default.
- W4281699648 hasConceptScore W4281699648C119857082 @default.
- W4281699648 hasConceptScore W4281699648C127413603 @default.
- W4281699648 hasConceptScore W4281699648C137293760 @default.
- W4281699648 hasConceptScore W4281699648C154945302 @default.
- W4281699648 hasConceptScore W4281699648C165801399 @default.
- W4281699648 hasConceptScore W4281699648C195324797 @default.
- W4281699648 hasConceptScore W4281699648C201995342 @default.
- W4281699648 hasConceptScore W4281699648C203005215 @default.
- W4281699648 hasConceptScore W4281699648C204321447 @default.
- W4281699648 hasConceptScore W4281699648C2780451532 @default.
- W4281699648 hasConceptScore W4281699648C39890363 @default.
- W4281699648 hasConceptScore W4281699648C41008148 @default.
- W4281699648 hasConceptScore W4281699648C44291984 @default.
- W4281699648 hasConceptScore W4281699648C50644808 @default.
- W4281699648 hasConceptScore W4281699648C622187 @default.
- W4281699648 hasConceptScore W4281699648C66322947 @default.
- W4281699648 hasConceptScore W4281699648C97931131 @default.
- W4281699648 hasIssue "3" @default.
- W4281699648 hasLocation W42816996481 @default.
- W4281699648 hasOpenAccess W4281699648 @default.
- W4281699648 hasPrimaryLocation W42816996481 @default.
- W4281699648 hasRelatedWork W2996854111 @default.
- W4281699648 hasRelatedWork W3033942572 @default.
- W4281699648 hasRelatedWork W3097571385 @default.
- W4281699648 hasRelatedWork W3155823939 @default.
- W4281699648 hasRelatedWork W3200578235 @default.
- W4281699648 hasRelatedWork W3212566403 @default.
- W4281699648 hasRelatedWork W4287761227 @default.
- W4281699648 hasRelatedWork W4310557741 @default.
- W4281699648 hasRelatedWork W4381786178 @default.
- W4281699648 hasRelatedWork W3045475294 @default.
- W4281699648 hasVolume "21" @default.
- W4281699648 isParatext "false" @default.
- W4281699648 isRetracted "false" @default.
- W4281699648 workType "article" @default.