Research

Paper

AI LLM February 19, 2026

Evaluating Extremely Low-Resource Machine Translation: A Comparative Study of ChrF++ and BLEU Metrics

Authors

Sanjeev Kumar, Preethi Jyothi, Pushpak Bhattacharyya

Abstract

Evaluating machine translation (MT) quality in extremely low-resource language (ELRL) scenarios poses unique challenges, as widely used metrics such as BLEU, effective in high-resource settings, often misrepresent quality in data-scarce contexts. This work presents a comparative analysis of BLEU, an n-gram-based metric, and ChrF++, a character-based metric, for MT evaluation in ELRL settings. We examine how each metric responds to translation artifacts, including hallucinations, repetition, source-text copying, and diacritic (\textit{matra}) variations across three ELRLs: Magahi, Bhojpuri, and Chhattisgarhi, with a focus on outputs from large language models (LLMs) and neural MT (NMT) systems. While recent work often relies solely on ChrF++, our findings show that BLEU, despite its lower absolute scores, provides complementary lexical-precision insights that improve interpretability.

Metadata

arXiv ID: 2602.17425
Provider: ARXIV
Primary Category: cs.CL
Published: 2026-02-19
Fetched: 2026-02-21 18:51

Related papers

Raw Data (Debug)
{
  "raw_xml": "<entry>\n    <id>http://arxiv.org/abs/2602.17425v1</id>\n    <title>Evaluating Extremely Low-Resource Machine Translation: A Comparative Study of ChrF++ and BLEU Metrics</title>\n    <updated>2026-02-19T14:56:42Z</updated>\n    <link href='https://arxiv.org/abs/2602.17425v1' rel='alternate' type='text/html'/>\n    <link href='https://arxiv.org/pdf/2602.17425v1' rel='related' title='pdf' type='application/pdf'/>\n    <summary>Evaluating machine translation (MT) quality in extremely low-resource language (ELRL) scenarios poses unique challenges, as widely used metrics such as BLEU, effective in high-resource settings, often misrepresent quality in data-scarce contexts. This work presents a comparative analysis of BLEU, an n-gram-based metric, and ChrF++, a character-based metric, for MT evaluation in ELRL settings. We examine how each metric responds to translation artifacts, including hallucinations, repetition, source-text copying, and diacritic (\\textit{matra}) variations across three ELRLs: Magahi, Bhojpuri, and Chhattisgarhi, with a focus on outputs from large language models (LLMs) and neural MT (NMT) systems. While recent work often relies solely on ChrF++, our findings show that BLEU, despite its lower absolute scores, provides complementary lexical-precision insights that improve interpretability.</summary>\n    <category scheme='http://arxiv.org/schemas/atom' term='cs.CL'/>\n    <published>2026-02-19T14:56:42Z</published>\n    <arxiv:comment>6 pages</arxiv:comment>\n    <arxiv:primary_category term='cs.CL'/>\n    <author>\n      <name>Sanjeev Kumar</name>\n    </author>\n    <author>\n      <name>Preethi Jyothi</name>\n    </author>\n    <author>\n      <name>Pushpak Bhattacharyya</name>\n    </author>\n  </entry>"
}