Sequential Fine-Tuning with Typologically Similar Languages for Yoruba Euphemism Detection

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 inclusion of intermediate typologically similar languages (e.g., Hausa, Wolof) in sequential fine-tuning impact cross-lingual euphemism detection performance in Yoruba compared to directly transferring from high-resource languages (e.g., English, Spanish), as measured by accuracy on XTREME-R and F1 score on SEED-Bench?

Autonomous synthesis report generated by Assignee Research. Tribunal consensus score: 8.7/10.