Predicting Human Translation Difficulty with Neural Machine Translation

2023/12/19 by Zheng Wei Lim, Ekaterina Vylomova, Lim, Zheng Wei +5
Computer Science · #Computation and Language (cs.CL) #FOS: Computer and information sciences #Natural Language Processing Techniques #Text Readability and Simplification

paper · pdf · doi:10.48550/arxiv.2312.11852

openalex publication_date 2023/12/19 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

Human translators linger on some words and phrases more than others, and predicting this variation is a step towards explaining the underlying cognitive processes. Using data from the CRITT Translation Process Research Database, we evaluate the extent to which surprisal and attentional features derived from a Neural Machine Translation (NMT) model account for reading and production times of human translators. We find that surprisal and attention are complementary predictors of translation difficulty, and that surprisal derived from a NMT model is the single most successful predictor of production duration. Our analyses draw on data from hundreds of translators operating across 13 language pairs, and represent the most comprehensive investigation of human translation difficulty to date.

Related

• Neural Machine Translation Quality and Post-Editing Performance
• Advancing Explainability in Neural Machine Translation: Analytical Metrics for Attention and Alignment Consistency
• Interrogating the Explanatory Power of Attention in Neural Machine\n Translation
• It's Easier to Translate out of English than into it: Measuring Neural Translation Difficulty by Cross-Mutual Information
• Neural Machine Translation with Recurrent Attention Modeling