Logo Lanfrica
  • Accueil
  • Atlas
  • Analyses
  • Documentation
  • Sign in

© 2026 Lanfrica. Tous droits réservés. Tous les droits d'auteur des ressources affichées sur le site Web Lanfrica appartiennent aux détenteurs de droits d'auteur d'origine, sauf indication contraire explicite.

SrcMix: Mixing of Related Source Languages Benefits Extremely Low-resource Machine Translation

Domaine:

natural language processing

Type de record:

paper
Créateur:
AssBhaJyoKum
Éditeur:
Und
Hôte:avatar
Multilingual models are widely used for machine translation (MT). However, their effectiveness for extremely low-resource languages (ELRLs) depends critically on how related languages are incorporated during fine-tuning. In this work, we study the role of language mixing directionality, linguistic relatedness, and script compatibility in ELRL translation. We propose SrcMix, a simple source-side mixing strategy that combines related ELRLs during fine-tuning while constraining the decoder to a single target language. Compared to its target-side counterpart TgtMix, SrcMix improves performance by +3 ChrF++ and +5 BLEU in high-resource to ELRL translations, and by +5 ChrF++ and +12 BLEU in mid-resource to ELRL translations. We also release the first Angika MT dataset and provide a systematic comparison of LLM (Aya-101) and NMT (mT5-Large) models under ELRL settings, highlighting the importance of directional mixing and linguistic compatibility.

Visit

doi.orgunderline.io

Tasks

machine translation

Tags

Computational LinguisticsNatural Language ProcessingArtificial Intelligence

Similaires

Translation or Recitation? Calibrating Evaluation Scores for Machine Translation of Extremely Low-Resource LanguagesNeural Machine Translation for Extremely Low-Resource African Languages: A Case Study on BambaraAddressing word-order Divergence in Multilingual Neural Machine Translation for extremely Low Resource LanguagesCharSpan: Utilizing Lexical Similarity to Enable Zero-Shot Machine Translation for Extremely Low-resource LanguagesSelectNoise: Unsupervised Noise Injection to Enable Zero-Shot Machine Translation for Extremely Low-resource LanguagesContinual Mixed-Language Pre-Training for Extremely Low-Resource Neural Machine Translation

Translation or Recitation? Calibrating Evaluation Scores for Machine Translation of Extremely Low-Resource Languages

The landscape of extremely low-resource machine translation (MT) is characterized by perplexing vari

Neural Machine Translation for Extremely Low-Resource African Languages: A Case Study on Bambara

Low-resource languages present unique challenges to (neural) machine translation. We discuss the cas

Addressing word-order Divergence in Multilingual Neural Machine Translation for extremely Low Resource Languages

Transfer learning approaches for Neural Machine Translation (NMT) train a NMT model on the assisting

CharSpan: Utilizing Lexical Similarity to Enable Zero-Shot Machine Translation for Extremely Low-resource Languages

We address the task of machine translation (MT) from extremely low-resource language (ELRL) to Engli

SelectNoise: Unsupervised Noise Injection to Enable Zero-Shot Machine Translation for Extremely Low-resource Languages

In this work, we focus on the task of machine translation (MT) from extremely low-resource language

Continual Mixed-Language Pre-Training for Extremely Low-Resource Neural Machine Translation

The data scarcity in low-resource languages has become a bottleneck to building robust neural machin