MTQE.en-he: Machine Translation Quality Estimation for English-Hebrew
This provides a benchmark for an under-resourced language pair, enabling future research, but it is incremental as it applies existing methods to new data.
The authors tackled the lack of a public benchmark for machine translation quality estimation in English-Hebrew by releasing MTQE.en-he, a dataset with 959 segments and human annotations, and showed that ensembling models improved performance by up to 6.4 percentage points in correlation metrics.
We release MTQE.en-he: to our knowledge, the first publicly available English-Hebrew benchmark for Machine Translation Quality Estimation. MTQE.en-he contains 959 English segments from WMT24++, each paired with a machine translation into Hebrew, and Direct Assessment scores of the translation quality annotated by three human experts. We benchmark ChatGPT prompting, TransQuest, and CometKiwi and show that ensembling the three models outperforms the best single model (CometKiwi) by 6.4 percentage points Pearson and 5.6 percentage points Spearman. Fine-tuning experiments with TransQuest and CometKiwi reveal that full-model updates are sensitive to overfitting and distribution collapse, yet parameter-efficient methods (LoRA, BitFit, and FTHead, i.e., fine-tuning only the classification head) train stably and yield improvements of 2-3 percentage points. MTQE.en-he and our experimental results enable future research on this under-resourced language pair.