Attitude Identification through Parameter-Efficient Fine-Tuning

Mariia Anisimova, Gabriella Lapesa, Sarka Zikanova


Abstract
We investigate automatic attitude detection in UN Security Council speeches using adapters. Following Martin and White’s Appraisal Theory, we identify three types of evaluative language: affect (emotional responses such as hope or concern), judgement (ethical evaluations of behavior), and appreciation (valuations of objects or situations). Training only 0.95% of BERT-large’s parameters, adapters achieve F1 scores ranging from 0.76 (affect) to 0.46 (appreciation), approaching full fine-tuning performance while enabling rapid task-specific experimentation. Differences in observed evaluation metrics mirror the pattern of the human inter-annotator agreement. This correlation suggests that computational difficulty reflects genuine linguistic ambiguity. Affect benefits from conventionalized diplomatic expressions, while appreciation faces context-dependent evaluation and severe class imbalance. Analysis demonstrates that evaluative intensity varies systematically across diplomatic contexts, with implications for corpus design in specialized discourse analysis.
Anthology ID:
2026.politicalnlp-1.22
Volume:
Proceedings of the 3rd Workshop on Natural Language Processing for Political Sciences (PoliticalNLP 2026)
Month:
May
Year:
2026
Address:
Palma, Mallorca (Spain)
Editors:
Haithem Afli, Houda Bouamor, Wajdi Zaghouani, Sahar Ghannay, Shehenaz Hossain
Venues:
PoliticalNLP | WS
SIG:
Publisher:
ELRA Language Resources Association (ELRA)
Note:
Pages:
204–209
Language:
External URL:
https://lrec.elra.info/lrec2026-ws-politicalnlp-22
DOI:
10.63317/34vqfemxyzgi
Bibkey:
Cite (ACL):
Mariia Anisimova, Gabriella Lapesa, and Sarka Zikanova. 2026. Attitude Identification through Parameter-Efficient Fine-Tuning. In Proceedings of the 3rd Workshop on Natural Language Processing for Political Sciences (PoliticalNLP 2026), pages 204–209, Palma, Mallorca (Spain). ELRA Language Resources Association (ELRA).
Cite (Informal):
Attitude Identification through Parameter-Efficient Fine-Tuning (Anisimova et al., PoliticalNLP 2026)
Copy Citation: