New AI method bypasses human annotation for machine translation error detection

Researchers have developed a new method for detecting errors in machine translation that does not require human annotation. This approach, called Iterative MBR Distillation, uses a large language model to generate its own training data, effectively creating pseudo-labels. Experiments show that models trained with this self-generated data perform better than those trained on human-annotated datasets, particularly at identifying specific error spans. AI

IMPACT This method could significantly reduce the cost and improve the consistency of training machine translation evaluation models.

RANK_REASON The cluster contains a research paper detailing a novel method for machine translation error span detection. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.AI →

• Boxuan Lyu
• Iterative MBR Distillation
• LLM
• Machine Translation
• WMT Metrics Shared Task

• paper
• model release

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →