Preserving affective nuance remains a challenge in Machine Translation (MT), where semantic equivalence often takes precedence over emotional fidelity. This paper evaluates the performance of three state-of-the-art Small Language Models (SLMs) -- EuroLLM, Aya Expanse, and Gemma -- in maintaining fine-grained emotions during backtranslation. Using the GoEmotions dataset, which comprises Reddit comments across 28 distinct categories, we assess emotional preservation across five European languages: German, French, Spanish, Italian, and Polish. Specifically, we investigate (i) the inherent capability of these SLMs to retain emotional sentiment, (ii) the efficacy of emotion-aware prompting in improving preservation, and (iii) the performance of ModernBERT as a contemporary alternative to BERT for emotion classification in MT evaluation.
@article{arxiv.2604.27920,
title = {Beyond Semantics: Measuring Fine-Grained Emotion Preservation in Small Language Model-Based Machine Translation},
author = {Dawid Wisniewski and Igor Czudy},
journal= {arXiv preprint arXiv:2604.27920},
year = {2026}
}