PulseExploreJournal ClubDebatesTrendingResearchersJournals
Instagram
HomeExploreJournal ClubTrending
Synapse
⌘+K
Synapse
April 10, 2026Translation Cognition & Behavior0 citationsOpen Access

Dependency-Based Syntactic Complexity and Post-Editing of GPT-4 Translations

How dependency-based syntactic complexity shapes post-editing of LLM-generated translations

View Full Paper

Authors

LZLonghui ZouMCMichaël CarlKent State UniversityJFJia FengRenmin University of China

Discussion

Loading...

Member takes

Overview

Examines the role of syntactic complexity in post-editing outcomes for English-to-Chinese translations, indicating that expertise moderates these effects.

Key Points

  • The aim is to explore how syntactic complexity affects the quality of post-editing in translations from GPT-4.
  • Data collected from 46 participants, including students and professionals.
  • Participants post-edited translations in Trados Studio.
  • Syntactic complexity operationalized using dependency-based metrics and English–Chinese specific indices.
  • Complexity in source texts and GPT-4 outputs predicts post-editing errors.
  • Experts demonstrate lower sensitivity to increasing complexity compared to students.
  • Termbase access decreases overall and terminology errors; advantage of experts is more pronounced in Light post-editing conditions.
Ask AI
Helpful
Bookmark
Share
View Full Paper

Cite This Study

Zou et al. (2026) studied this question.

synapsesocial.com/papers/69d895ea6c1944d70ce07068https://doi.org/10.1075/tcb.00100.zou
Ask AI
Helpful
Bookmark
Share
View Full Paper