Multilingual Knowledge Transfer under Data Constraints via Lexical Interventions

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Sedova, Anastasiia, Schluter, Natalie, Seto, Skyler, ter Hoeve, Maartje
Formato: Preprint
Publicado: 2026
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866917525452750848
author Sedova, Anastasiia
Schluter, Natalie
Seto, Skyler
ter Hoeve, Maartje
author_facet Sedova, Anastasiia
Schluter, Natalie
Seto, Skyler
ter Hoeve, Maartje
contents Cross-lingual knowledge transfer is critical for building high-performing multilingual language models for languages with insufficient training data. When target language data is scarce, the knowledge required for many downstream tasks involving scientific reasoning, commonsense inference, and world knowledge must be acquired primarily from the high-resource language, making effective knowledge transfer essential. Existing methods for improving such cross-lingual knowledge transfer require large amounts of parallel data, translation systems, auxiliary models, or additional training stages that are largely unavailable for many languages. We propose LINK - a data-level intervention method that improves knowledge transfer during model pretraining through lexical substitutions in high-resource part of pretraining data using bilingual vocabularies. For a given replacement ratio, randomly selected words in a portion of the high-resource (English) training corpus are swapped with their word-level translations, requiring no additional model training and only a bilingual vocabulary, which can be obtained at near-zero cost for virtually any language. Evaluation on eight languages across five model sizes shows notable improvements on downstream tasks in the target language, with up to a 2x speedup in training to reach equivalent performance.
format Preprint
id arxiv_https___arxiv_org_abs_2605_23885
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Multilingual Knowledge Transfer under Data Constraints via Lexical Interventions
Sedova, Anastasiia
Schluter, Natalie
Seto, Skyler
ter Hoeve, Maartje
Computation and Language
Cross-lingual knowledge transfer is critical for building high-performing multilingual language models for languages with insufficient training data. When target language data is scarce, the knowledge required for many downstream tasks involving scientific reasoning, commonsense inference, and world knowledge must be acquired primarily from the high-resource language, making effective knowledge transfer essential. Existing methods for improving such cross-lingual knowledge transfer require large amounts of parallel data, translation systems, auxiliary models, or additional training stages that are largely unavailable for many languages. We propose LINK - a data-level intervention method that improves knowledge transfer during model pretraining through lexical substitutions in high-resource part of pretraining data using bilingual vocabularies. For a given replacement ratio, randomly selected words in a portion of the high-resource (English) training corpus are swapped with their word-level translations, requiring no additional model training and only a bilingual vocabulary, which can be obtained at near-zero cost for virtually any language. Evaluation on eight languages across five model sizes shows notable improvements on downstream tasks in the target language, with up to a 2x speedup in training to reach equivalent performance.
title Multilingual Knowledge Transfer under Data Constraints via Lexical Interventions
topic Computation and Language
url https://arxiv.org/abs/2605.23885