20.500.12556/RUL-178163 Measuring catastrophic forgetting in cross-lingual classification transfer paradigms and tuning strategies Cross-lingual transfer leverages knowledge from a resource-rich source language, commonly English, to enhance performance in less-resourced target languages. Two widely used strategies are: Cross-Lingual Validation (CLV), which involves training on the source language and validating on the target language, and Intermediate Training (IT), where models are first fine-tuned on the source language and then further trained on the target language. While both strategies have been studied, their effects on encoder-based models for classification tasks remain underexplored. In this paper, we systematically compare these strategies across six multilingual classification tasks, evaluating downstream performance, catastrophic forgetting, and both zero-shot and full-shot scenarios. Additionally, we contrast parameter-efficient adapter methods with full-parameter fine-tuning. Our results show that IT generally performs better in the target language, whereas CLV more effectively preserves source-language knowledge across multiple cross-lingual transfers. These findings underscore the trade-offs between optimizing target performance and mitigating catastrophic forgetting. cross-lingual transfer cross-lingual learning catastrophic-forgetting document classification medjezikovno učenje medjezikovni prenos globoko učenje strojno učenje true false true Angleški jezik Slovenski jezik Članek v reviji 2026-01-20 14:45:51 2026-01-20 14:45:54 2026-01-21 04:18:06 0000-00-00 00:00:00 2025 0 0 Str. 33509-33520 Vol. 13 2025 0000-00-00 Zaloznikova Objavljeno NiDoloceno 0000-00-00 0000-00-00 0000-00-00 004.8 2169-3536 10.1109/ACCESS.2025.3543608 227115523 Measuring_Catastrophic_Forgetting_in_Cross-Lingual_Classification_Transfer_Paradigms_and_Tuning_Strategies.pdf Measuring_Catastrophic_Forgetting_in_Cross-Lingual_Classification_Transfer_Paradigms_and_Tuning_Strategies.pdf 1 45A7548CE216F471D14CF068A4431CA2 540826a16f77625812d65603938f20e3c7698e84990c830ffa272be75b0bb531 b9de83da-f606-11f0-bf92-0050569b8976 https://repozitorij.uni-lj.si/Dokument.php?lang=slv&id=225755 https://ieeexplore.ieee.org/document/10892119 1 48e96475-f606-11f0-bf92-0050569b8976 https://repozitorij.uni-lj.si/Dokument.php?lang=slv&id=225754 Fakulteta za računalništvo in informatiko 0 0 0