DimStance: Multilingual Datasets for Dimensional Stance Analysis

TL;DR

DimStance introduces a multilingual dataset with valence-arousal annotations for nuanced stance analysis, enabling fine-grained, emotion-aware stance detection across languages and domains.

Contribution

It presents the first dimensional stance resource with VA annotations, covering multiple languages and domains, and benchmarks models for VA prediction.

Findings

Fine-tuned LLM regressors perform competitively.

Challenges remain in low-resource languages.

Token-based generation has limitations.

Abstract

Stance detection is an established task that classifies an author's attitude toward a specific target into categories such as Favor, Neutral, and Against. Beyond categorical stance labels, we leverage a long-established affective science framework to model stance along real-valued dimensions of valence (negative-positive) and arousal (calm-active). This dimensional approach captures nuanced affective states underlying stance expressions, enabling fine-grained stance analysis. To this end, we introduce DimStance, the first dimensional stance resource with valence-arousal (VA) annotations. This resource comprises 11,746 target aspects in 7,365 texts across five languages (English, German, Chinese, Nigerian Pidgin, and Swahili) and two domains (politics and environmental protection). To facilitate the evaluation of stance VA prediction, we formulate the dimensional stance regression task, analyze cross-lingual VA patterns, and benchmark pretrained and large language models under regression and prompting settings. Results show competitive performance of fine-tuned LLM regressors, persistent challenges in low-resource languages, and limitations of token-based generation. DimStance provides a foundation for multilingual, emotion-aware, stance analysis and benchmarking.