Towards abstractive summarization in Hungarian

Makrai Márton and Tündik Máté Ákos and Indig Balázs and Szaszák György: Towards abstractive summarization in Hungarian.

Cikk, tanulmány, mű msznykonf_018_505-519.pdf Download (639kB)

Abstract

We publish an abstractive summarizer for Hungarian, an encoder-decoder model initialized with huBERT, and fine-tuned on the ELTE.DH corpus of former Hungarian news portals. The model produces fluent output in the correct topic, but it hallucinates frequently. Our quantitative evaluation on automatic and human transcripts of news (with automatic and human-made punctuation) shows that the model is robust with respect to errors in either automatic speech recognition or automatic punctuation restoration.

Item Type:	Conference or Workshop Item
Heading title:	Poszter, laptopos bemutató
Journal or Publication Title:	Magyar Számítógépes Nyelvészeti Konferencia
Date:	2022
Volume:	18
ISBN:	978-963-306-848-9
Page Range:	pp. 505-519
Language:	English
Place of Publication:	Szeged
Event Title:	Magyar számítógépes nyelvészeti konferencia (18.) (2022) (Szeged)
Related URLs:	http://acta.bibl.u-szeged.hu/75797/
Uncontrolled Keywords:	Nyelvészet - számítógép alkalmazása
Additional Information:	Bibliogr.: p. 516-519. ; ill. ; összefoglalás angol nyelven
Subjects:	01. Natural sciences 01. Natural sciences > 01.02. Computer and information sciences 06. Humanities 06. Humanities > 06.02. Languages and Literature
Date Deposited:	2022. May. 25. 13:31
Last Modified:	2022. Nov. 08. 11:49
URI:	http://acta.bibl.u-szeged.hu/id/eprint/75896

Actions (login required)

View Item