HomeLREC 2026WorkshopsPOLITICALNLPlrec2026-ws-politicalnlp-22
Back to POLITICALNLP 2026
LREC 2026workshop

Attitude Identification through Parameter-Efficient Fine-Tuning

Proceedings of the Second Workshop on Building Educational Applications Using NLP

DOI:10.63317/34vqfemxyzgi

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.

Details

Paper ID
lrec2026-ws-politicalnlp-22
Pages
pp. 204-209
BibKey
anisimova-etal-2026-attitude
Editors
N/A
Publisher
European Language Resources Association (ELRA)
ISSN
N/A
ISBN
N/A
Workshop
Proceedings of the Second Workshop on Building Educational Applications Using NLP
Location
Palma, Mallorca, Spain
Date
11 - 16 May 2026

Authors

  • MA

    Mariia Anisimova

  • GL

    Gabriella Lapesa

  • SZ

    Sarka Zikanova

Links