| Title |
Analyzing the effects of domain-adaptive pretraining on propaganda technique detection in a low-resource language |
| Authors |
Rizgelienė, Ieva ; Seeck, Hannele |
| DOI |
10.1109/TAI.2026.3723357 |
| Full Text |
|
| Is Part of |
IEEE Transactions on artificial intelligence.. New York : IEEE. 2026, Early Access, p. 1-15.. eISSN 2691-4581 |
| Keywords [eng] |
domain-adaptive pretraining ; propaganda techniques ; low-resource language ; transformers ; NLP ; masked language |
| Abstract [eng] |
This study presents one of the first systematic investigations of domain-adaptive pretraining for propaganda technique detection in Lithuanian. We perform masked language modeling-based domain-adaptive pretraining on a corpus of 63,648 news articles from non-credible outlets. We evaluate its impact using masked-language probing, propaganda technique recognition through sequence tagging and sentence classification, and comparison with a large language model in a zero-shot setting. In addition to quantitative evaluation, we conduct a qualitative analysis based on a case study of annotation behavior. The results show that domain-adaptive pretraining improves propaganda technique detection in Lithuanian, enhances the ranking of contextually meaningful terms, and improves performance for most techniques, with gains varying across tasks. |
| Published |
New York : IEEE |
| Type |
Journal article |
| Language |
English |
| Publication date |
2026 |
| CC license |
|
