@inproceedings{MTMT:34560602, title = {Személyes adatok azonosítása és automatikus lecserélése magyar nyelvű szövegekben}, url = {https://m2.mtmt.hu/api/publication/34560602}, author = {Novák, Attila and Novák, Borbála}, booktitle = {XX. Magyar Számítógépes Nyelvészeti Konferencia}, unique-id = {34560602}, year = {2024}, pages = {117-129} } @inproceedings{MTMT:34531850, title = {SHunQA: egy nyíltkérdés-megválaszoló rendszer}, url = {https://m2.mtmt.hu/api/publication/34531850}, author = {Berkecz, Péter and Zombori, Tamás and Banga, Gergő and Szabó, Gergő and Szántó, Zsolt and Novák, Attila and Farkas, Richárd}, booktitle = {XX. Magyar Számítógépes Nyelvészeti Konferencia}, unique-id = {34531850}, year = {2024}, pages = {73-84} } @inproceedings{MTMT:34161870, title = {A Question Answering Benchmark Database for Hungarian}, url = {https://m2.mtmt.hu/api/publication/34161870}, author = {Novák, Attila and Novák, Borbála and Zombori, Tamás and Szabó, Gergő and Szántó, Zsolt and Farkas, Richárd}, booktitle = {Proceedings of the 17th Linguistic Annotation Workshop (LAW-XVII)}, doi = {10.18653/v1/2023.law-1.19}, unique-id = {34161870}, year = {2023}, pages = {188-198} } @inproceedings{MTMT:34158531, title = {A Pseudonymization Prototype for Hungarian}, url = {https://m2.mtmt.hu/api/publication/34158531}, author = {Novák, Attila and Novák, Borbála}, booktitle = {12th Symposium on Languages, Applications and Technologies (SLATE 2023)}, doi = {10.4230/OASIcs.SLATE.2023.3}, unique-id = {34158531}, year = {2023}, pages = {3:1-3:10} } @inproceedings{MTMT:34158455, title = {Identification of Lemmatization Errors Using Neural Models}, url = {https://m2.mtmt.hu/api/publication/34158455}, author = {Novák, Attila and Novák, Borbála}, booktitle = {Computational Linguistics and Intelligent Text Processing}, doi = {10.1007/978-3-031-23793-5_32}, unique-id = {34158455}, year = {2023}, pages = {399-407} } @{MTMT:33675492, title = {POS, ANA and LEM: Word Embeddings Built from Annotated Corpora Perform Better (Best Paper Award, Second Place)}, url = {https://m2.mtmt.hu/api/publication/33675492}, author = {Novák, Attila and Novák, Borbála}, booktitle = {Computational Linguistics and Intelligent Text Processing}, doi = {10.1007/978-3-031-23793-5_29}, unique-id = {33675492}, year = {2023}, pages = {360-370} } @article{MTMT:33624221, title = {Fine-tuning and multilingual pre-training for abstractive summarization task for the Arabic language}, url = {https://m2.mtmt.hu/api/publication/33624221}, author = {Kahla, Mram and Novák, Attila and Yang, Zijian Győző}, doi = {10.33039/ami.2022.11.002}, journal-iso = {ANN MATH INFORM}, journal = {ANNALES MATHEMATICAE ET INFORMATICAE}, volume = {57}, unique-id = {33624221}, issn = {1787-5021}, year = {2023}, eissn = {1787-6117}, pages = {24-35} } @inproceedings{MTMT:33614189, title = {MILQA kérdés-válasz benchmark adatbázis}, url = {https://m2.mtmt.hu/api/publication/33614189}, author = {Novák, Attila and Novák, Borbála}, booktitle = {XIX. Magyar Számítógépes Nyelvészeti Konferencia, MSZNY-2023}, unique-id = {33614189}, year = {2023}, pages = {203-216} } @article{MTMT:33287822, title = {Cross-lingual transfer of knowledge in distributional language models: Experiments in Hungarian}, url = {https://m2.mtmt.hu/api/publication/33287822}, author = {Novák, Attila and Novák, Borbála}, doi = {10.1556/2062.2022.00580}, journal-iso = {ACTA LING ACAD}, journal = {ACTA LINGUISTICA ACADEMICA}, volume = {69}, unique-id = {33287822}, issn = {2559-8201}, abstract = {In this paper, we argue that the very convincing performance of recent deep-neural-model-based NLP applications has demonstrated that the distributionalist approach to language description has proven to be more successful than the earlier subtle rule-based models created by the generative school. The now ubiquitous neural models can naturally handle ambiguity and achieve human-like linguistic performance with most of their training consisting only of noisy raw linguistic data without any multimodal grounding or external supervision refuting Chomsky's argument that some generic neural architecture cannot arrive at the linguistic performance exhibited by humans given the limited input available to children. In addition, we demonstrate in experiments with Hungarian as the target language that the shared internal representations in multilingually trained versions of these models make them able to transfer specific linguistic skills, including structured annotation skills, from one language to another remarkably efficiently.}, year = {2022}, eissn = {2560-1016}, pages = {405-449} } @inproceedings{MTMT:33118689, title = {NerKor+Cars-OntoNotes++}, url = {https://m2.mtmt.hu/api/publication/33118689}, author = {Novák, Attila and Novák, Borbála}, booktitle = {LREC 2022, Thirteeth International Conference on Language Resources and Evaluation}, unique-id = {33118689}, year = {2022}, pages = {1907-1916} }