Researchers have introduced CzechDocs, a new dataset designed for evaluating machine translation systems that preserve document formatting. This multiway parallel dataset includes documents in Czech and several minority languages such as Ukrainian, English, and Vietnamese, formatted in HTML, DOCX, and PDF. The dataset aims to advance research in document-level translation, with a validation split and evaluation toolkit already released, and a test split planned for a future shared task. AI
IMPACT Facilitates the development of machine translation systems capable of maintaining document formatting, crucial for accurate cross-lingual information retrieval.
RANK_REASON The cluster describes a new dataset released on arXiv for research purposes.
AI-generated summary · Google Gemini · from 2 sources. How we write summaries →
