When Does Language Transfer Help? Sequential Fine-Tuning for Cross-Lingual Euphemism Detection

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Sammartino, Julia, Barak, Libby, Peng, Jing, Feldman, Anna
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916902379454464
author Sammartino, Julia
Barak, Libby
Peng, Jing
Feldman, Anna
author_facet Sammartino, Julia
Barak, Libby
Peng, Jing
Feldman, Anna
contents 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 performance, especially for low-resource languages like Yoruba and Turkish. XLM-R achieves larger gains but is more sensitive to pretraining gaps and catastrophic forgetting, while mBERT yields more stable, though lower, results. These findings highlight sequential fine-tuning as a simple yet effective strategy for improving euphemism detection in multilingual models, particularly when low-resource languages are involved.
format Preprint
id arxiv_https___arxiv_org_abs_2508_11831
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle When Does Language Transfer Help? Sequential Fine-Tuning for Cross-Lingual Euphemism Detection
Sammartino, Julia
Barak, Libby
Peng, Jing
Feldman, Anna
Computation and Language
Artificial Intelligence
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 performance, especially for low-resource languages like Yoruba and Turkish. XLM-R achieves larger gains but is more sensitive to pretraining gaps and catastrophic forgetting, while mBERT yields more stable, though lower, results. These findings highlight sequential fine-tuning as a simple yet effective strategy for improving euphemism detection in multilingual models, particularly when low-resource languages are involved.
title When Does Language Transfer Help? Sequential Fine-Tuning for Cross-Lingual Euphemism Detection
topic Computation and Language
Artificial Intelligence
url https://arxiv.org/abs/2508.11831