Measuring the Cost of Variety Conflation in Multilingual MT Evaluation: Adding Mozambican Xichangana, Nyanja and Sena to FLORES+

arXiv:2609.13847v1 Announce Type: new Abstract: In this paper, we extend FLORES+ with Portuguese-source evaluation sets for three Mozambican Bantu varieties: Xichangana, Mozambican Nyanja, and Sena. We compare Xichangana with the existing Tsonga reference and Mozambican Nyanja with Chichewa, and evaluate NLLB-200, Google Translate, GPT, and a variant-aware NLLB model. Holding system output fixed reveals substantial reference sensitivity. On extit{devtest}, changing only the reference from Tsonga to Xichangana reduces spBLEU by 13.10 points for NLLB-200 and 15.30 for Google. On matched Nyanja subsets, replacing Chichewa with Mozambican Nyanja produces smaller but consistent reductions of 3.03 and 6.10 spBLEU, respectively. Variant-aware fine-tuning reverses this pattern on the intended targets: relative to NLLB-200, it improves Xichangana by 7.04 spBLEU and Mozambican Nyanja by 5.33 on extit{devtest}, while losing performance on the sibling references. GPT is competitive on Tsonga and Chichewa but substantially weaker on the Mozambican varieties. For Sena, the finetuned model reaches 12.64 spBLEU and 36.21 chrF++ on extit{devtest}. These findings motivate variety-aware language identifiers, references, and reporting for cross-border languages or language dialects/variants. The data is publicly available on Hugging Face at https://huggingface.co/datasets/MOZNLP/FLORES_MOZ

Related

• Harnessing Linguistic Dissimilarity for Language Generalization on Unseen Low-Resource Varieties
• Apples to Apples? Towards Comparable Crosslingual Language Model Evaluation
• HPLT 3.0: Very Large-Scale Multilingual Resources for LLMs and MT. Mono- and Bi-lingual Data, Multilingual Evaluation, and Pre-Trained Models
• When the Knowledge Base Becomes the Gold Standard: Measuring Resource-Shared Evaluation Loops in Entity-Level Machine Translation

Source: arXiv cs.CL | 2026-09-15