Research links translationese to cognitive load in translation tasks

A new research paper proposes that "translationese," the systematic divergence of translated texts from original target-language works, can be explained by the cognitive load inherent in the translation task itself. Researchers operationalized translationese using an automatic classifier and measured task difficulty through LLM surprisal, syntactic complexity, and semantic alternatives. Their findings, primarily from an English-German corpus, indicate that translation task difficulty, particularly cross-lingual transfer, is a significant predictor of translationese, with source-text syntactic complexity and translation-solution entropy being the strongest indicators. AI

IMPACT This research offers a new perspective on language processing and potential biases in machine translation systems.

RANK_REASON The cluster contains a research paper published on arXiv detailing a new linguistic theory. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.CL →

• alphaXiv
• arXiv
• CatalyzeX
• DagsHub
• English-German School
• Gotit.pub
• Hugging Face
• Maria Kunilovskaya
• ScienceCast
• translationese

• paper
• other

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