Skip to content
conftrace_
2026 EMNLP EMNLP 2026

Expanding Chechen Machine Translation Resources: SMOL, BOUQuET, and WMT24++ Extensions

AbstractWe introduce a new fork of the SMOL training dataset along with forks of the BOUQuET and WMT24++ translation benchmarks for the Chechen language. We train a model on SMOL and calculate BLEU and chrF++ scores on the introduced benchmarks to evaluate how training on such a limited amount of data may contribute to the model’s Chechen-Russian translation capability. The introduced resources provide a unified evaluation framework for Chechen machine translation and allow measuring the transfer of improvements across different benchmark datasets. Our experiments establish baseline results for future research on Chechen translation and demonstrate the potential of utilizing small-scale parallel corpora for improving low-resource language models.

🌉 Interdisciplinary Bridge - Artificial Intelligence and Interdisciplinary and Natural Language Processing

More from EMNLP 2026

All EMNLP 2026 papers →