Experts, Errors, and Context: A Large-Scale Study of Human Evaluation for Machine Translation

Abstract

Human evaluation of modern high-quality machine translation systems is a difficult problem, and there is increasing evidence that inadequate evaluation procedures can lead to erroneous conclusions. While there has been considerable research on human evaluation, the field still lacks a commonly-accepted standard procedure. As a step toward this goal, we propose an evaluation methodology grounded in explicit error analysis, based on the Multidimensional Quality Metrics (MQM) framework. We carry out the largest MQM research study to date, scoring the outputs of top systems from the WMT 2020 shared task in two language pairs using annotations provided by professional translators with access to full document context. We analyze the resulting data extensively, finding among other results a substantially different ranking of evaluated systems from the one established by the WMT crowd workers, exhibiting a clear preference for human over machine output. Surprisingly, we also find that automatic metrics based on pre-trained embeddings can outperform human crowd workers. We make our corpus publicly available for further research.

References in corpus (2)

• A Set of Recommendations for Assessing Human-Machine Parity in Language Translation
• Translationese in Machine Translation Evaluation

Cited by in corpus (13)

• To Ship or Not to Ship: An Extensive Evaluation of Automatic Metrics for Machine Translation
• adaptMLLM: Fine-Tuning Multilingual Language Models on Low-Resource Languages with Integrated LLM Playgrounds
• Human Evaluation of English--Irish Transformer-Based NMT
• Angler: Helping Machine Translation Practitioners Prioritize Model Improvements
• Uncertainty-Aware Machine Translation Evaluation
• USCORE: An Effective Approach to Fully Unsupervised Evaluation Metrics for Machine Translation
• DivEMT: Neural Machine Translation Post-Editing Effort Across Typologically Diverse Languages
• Regressive Ensemble for Machine Translation Quality Evaluation
• cushLEPOR: customising hLEPOR metric using Optuna for higher agreement with human judgments or pre-trained language model LaBSE
• QE4PE: Word-level Quality Estimation for Human Post-Editing
• The statistical advantage of automatic NLG metrics at the system level
• Natural Language Processing RELIES on Linguistics
• The Eval4NLP Shared Task on Explainable Quality Estimation: Overview and Results