Поиск :
Личный кабинет :
Электронный каталог: Ilina, A. - Comparative Analysis and Evaluation of the Consistency of Methods for Integrating Semantic Repres...
Ilina, A. - Comparative Analysis and Evaluation of the Consistency of Methods for Integrating Semantic Repres...

Статья
Автор: Ilina, A.
Физика элементарных частиц и атомного ядра: Comparative Analysis and Evaluation of the Consistency of Methods for Integrating Semantic Repres... : Abstract
б.г.
ISBN отсутствует
Автор: Ilina, A.
Физика элементарных частиц и атомного ядра: Comparative Analysis and Evaluation of the Consistency of Methods for Integrating Semantic Repres... : Abstract
б.г.
ISBN отсутствует
Статья
Ilina, A.
Comparative Analysis and Evaluation of the Consistency of Methods for Integrating Semantic Representations of Texts : Abstract / A.Ilina, P.Zrelov // Физика элементарных частиц и атомного ядра. – 2026. – Т. 57, № 4. – P. 793. – URL: https://www1.jinr.ru/Pepan/v-57-4/Ilina_ann.pdf.
We explore methods for aggregating the results of three embedding models to estimate semantic similarity of texts: cosine similarities averaging, vector concatenation, and selection of one of the three cosine similarities based on principal component analysis and singular value decomposition methods. Statistical analysis showed that averaging cosine proximity as an aggregation method demonstrates the best results. The obtained results can be used to improve the robustness of models for semantic proximity evaluation of texts when using embedding ensembles.
Спец.(статьи,препринты) = Ц 849 - Искусственный интеллект. Теория и практика
Спец.(статьи,препринты) = Ц 840 г1 - Банки и базы данных
ОИЯИ = ОИЯИ (JINR)2026
Ilina, A.
Comparative Analysis and Evaluation of the Consistency of Methods for Integrating Semantic Representations of Texts : Abstract / A.Ilina, P.Zrelov // Физика элементарных частиц и атомного ядра. – 2026. – Т. 57, № 4. – P. 793. – URL: https://www1.jinr.ru/Pepan/v-57-4/Ilina_ann.pdf.
We explore methods for aggregating the results of three embedding models to estimate semantic similarity of texts: cosine similarities averaging, vector concatenation, and selection of one of the three cosine similarities based on principal component analysis and singular value decomposition methods. Statistical analysis showed that averaging cosine proximity as an aggregation method demonstrates the best results. The obtained results can be used to improve the robustness of models for semantic proximity evaluation of texts when using embedding ensembles.
Спец.(статьи,препринты) = Ц 849 - Искусственный интеллект. Теория и практика
Спец.(статьи,препринты) = Ц 840 г1 - Банки и базы данных
ОИЯИ = ОИЯИ (JINR)2026
На полку