Low-Resource MT

MT: Machine Translation

Overview

The field studies how to translate languages that lack the large parallel-text collections and language technologies available for widely used languages, helping broaden access to information and services. Common approaches include building curated corpora, adapting multilingual models such as NLLB and mT5, transferring knowledge from related languages, and creating synthetic training data through back-translation or language-model generation. Researchers also explore retrieval-based prompting, linguistic resources such as dictionaries and grammars, and reinforcement learning with semantic or quality-estimation rewards. Progress remains uneven: results depend heavily on data quality, language variety, script and tokenization, while metric reliability, data contamination, and the gap between automatic scores and human judgment remain open concerns.

Latest papers

95 papers7 papers in the last four weeks, down 53% on the four weeks before. 0.1% of all new papers.

Feb 17, 2026

LuxMT Technical Report

We introduce LuxMT, a machine translation system based on Gemma 3 27B and fine-tuned for translation from Luxembourgish (LB) into French (FR) and English (EN). To assess translation performance, we construct a novel benchmark covering LB-FR, LB-EN, and LB-FR using human-translated data from Luci, a tourist magazine about Luxembourg. Training data stems from LuxAlign, a parallel corpus of multilingual Luxembourgish news articles, and LB parliamentary transcripts augmented with Google Translate. We filter the data using LuxEmbedder, LB sentence embeddings, to remove low-equivalence segment-pairs. Overall, LuxMT's results suggest strong improvements over the Gemma 3 baseline, even for translating LB to German (DE), despite the training data not containing any DE. We also explore LuxEmbedder's potential to be used as a quality estimation metric and find strong correlations with other reference-based metrics. However, we call for further research to fully assess the metric's utility and advise using it with caution.
Feb 16, 2026

Crowdsourcing Piedmontese to Test LLMs on Non-Standard Orthography

We present a crowdsourced dataset for Piedmontese, an endangered Romance language of northwestern Italy. The dataset comprises 145 Italian-Piedmontese parallel sentences derived from Flores+, with translations produced by speakers writing in their natural orthographic style rather than adhering to standardized conventions, along with manual word alignment. We use this resource to benchmark several large language models on tokenization parity, topic classification, and machine translation. Our analysis reveals that Piedmontese incurs a tokenization penalty relative to higher-resource Romance languages, yet LLMs achieve classification performance approaching that of Italian, French, and English. Machine translation results are asymmetric: models translate adequately from Piedmontese into high-resource languages, but generation into Piedmontese remains challenging. The dataset and code are publicly released.
Jan 15, 2026

Is MT Ready for the Next Crisis or Pandemic?

Communication in times of crisis is essential. However, there is often a mismatch between the language of governments, aid providers, doctors, and those to whom they are providing aid. It is reasonable to turn to commercial MT tools in these scenarios. But how effective are these tools for translating to and from low resource languages, particularly in the crisis or medical domain? In this study, we evaluate 5 commercial systems used for MT (Google Translate, Microsoft Translator, GPT, Gemini and Claude) against the TICO-19 dataset, which contains pandemic-related sentences from a large set of high-priority languages spoken by communities most likely to be adversely affected in the next pandemic. We assess the current degree of "readiness" for this next pandemic based on the usability of the output translations.
Dec 3, 2025

Adapting Large Language Models to Low-Resource Tibetan: A Two-Stage Continual and Supervised Fine-Tuning Study

Adapting large language models (LLMs) to low-resource languages remains a major challenge due to data scarcity and cross-lingual drift. This work presents a two-stage adaptation of Qwen2.5-3B to Tibetan, a morphologically rich and underrepresented language. We employ Continual Pretraining (CPT) to establish Tibetan linguistic grounding, followed by Supervised Fine-Tuning (SFT) for task and translation specialization. Empirical evaluations demonstrate a consistent decrease in perplexity (from 2.98 →\rightarrow 1.54) and substantial improvements in Chinese→\rightarrowTibetan translation quality (BLEU: 0.046 →\rightarrow 0.261; chrF: 2.2 →\rightarrow 6.6). Layer-wise analysis across 435 layers in Qwen3-4B reveals that adaptation primarily concentrates on embedding and output heads, with mid--late MLP projections encoding domain-specific transformations. Our findings suggest that CPT constructs a Tibetan semantic manifold while SFT sharpens task alignment with minimal representational disruption. This study provides the first quantitative exploration of Tibetan adaptation dynamics for LLMs, and offers an open, reproducible framework for extending multilingual foundation models to low-resource settings.
Apr 3, 2025

Breaking the Silence: A Dataset and Benchmark for Bangla Text-to-Gloss Translation

Gloss is a written approximation that bridges Sign Language (SL) and its corresponding spoken language. Despite a deaf and hard-of-hearing population of at least 3 million in Bangladesh, Bangla Sign Language (BdSL) remains largely understudied, with very limited prior work on Bangla text-to-gloss translation and no publicly available parallel dataset at the time of this work. To address this gap, we construct the first publicly available Bangla text-to-gloss dataset, consisting of 1,000 manually annotated and 4,000 synthetically generated Bangla sentence-gloss pairs, along with 159 expert human-annotated pairs used as a test set. Our experimental framework performs a comparative analysis between several fine-tuned open-source models and a leading closed-source LLM to evaluate their performance in low-resource BdSL translation. GPT-5.4 achieves the best automatic metric scores, while open-source models perform competitively despite being much smaller. This work introduces the first publicly available expert-annotated Bangla text-to-gloss dataset, benchmark and trained model for Bangla text-to-gloss translation. It also demonstrates the effectiveness of systematically generated synthetic data for addressing challenges in low-resource sign language translation.