BlameBERT Classifier Identifies Political Blame Trends in Danish Parliament
September 21, 2026
BlameBERT achieves an F1 score of 0.80 for detecting political blame attribution in low-to-mid resource languages. The study uses this classifier to show a significant increase in parliamentary blame from 2019 to 2026, noting that opposition parties attribute blame more frequently than governing parties.
HOW THIS AFFECTS YOU
●
researcherYou can use this annotation-efficient pipeline for blame attribution tasks in similar low-resource linguistic contexts.