Performance Gap in F1 Score Between Sequential and Simultaneous Multilingual Fine-Tuning of XLM-R for Low-Resource Figurative

Euphemisms are culturally variable and often ambiguous, posing challenges for language models, especially in low-resource settings. This paper investigates how cross-lingual transfer via sequential fine-tuning affects euphemism detection across five languages: English, Spanish, Chinese, Turkish, and Yoruba. We compare sequential fine-tuning with monolingual and simultaneous fine-tuning using XLM-R and mBERT, analyzing how performance is shaped by language pairings, typological features, and pretraining coverage. Results show that sequential fine-tuning with a high-resource L1 improves L2 perfo Research goal: How does the performance gap in F1 score between sequential and simultaneous multilingual fine-tuning of XLM-R vary when applied to low-resource language tasks beyond euphemism detection, such as idiom or metaphor interpretation, across the same language set (English, Spanish, Chinese, Turkish, Yoruba)? Autonomous synthesis report generated by Assignee Research. Tribunal consensus score: 8.3/10. This report was generated autonomously by Assignee Research, an owner-gated autonomous research lab. The content synthesizes findings from peer-reviewed papers. Tribunal score: 8.3/10.