Emotion Profiling in LLM-Based Literary Translation: Systematic Shifts Across MT and Post-Editing
Quick Answer
This study analyzes emotional profiles in LLM translations of Atwood's 'Oryx and Crake' and their post-edited versions, revealing that MT systems create distinct emotional fingerprints, which compromise the preservation of the author's voice.
Quick Take
Using a multilingual approach, the research highlights significant emotional shifts post-editing compared to human translations.
Key Points
- translations exhibit identifiable emotional profiles compared to human translations.
- Post-editing reshapes emotional content toward human-like norms.
- Distinct emotional fingerprints are observed across different MT systems.
- The study uses a large-scale corpus of contemporary Italian science fiction.
- Limited preservation of author's voice noted in MT translations.
Paper Resources
📖 Reader Mode
~1 min readAbstract:This paper investigates whether LLM translations exhibit identifiable emotional profiles and how post-editing reshapes them toward human-like norms. We compare LLM translations of Margaret Atwood's Oryx and Crake with their post-edited versions and a human translation, using a large-scale corpus of contemporary Italian science-fiction as a baseline. We examine emotion through lexicon-based and multilingual modeling, conducting a fine-grained analysis of emotional variation across systems. We find that MT systems introduce model-specific and statistically significant emotional fingerprints across translations, leading to a limited preservation of an author's voice.
| Subjects: | Computation and Language (cs.CL); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2606.10113 [cs.CL] |
| (or arXiv:2606.10113v1 [cs.CL] for this version) | |
| https://doi.org/10.48550/arXiv.2606.10113 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Antonio Castaldo [view email]
[v1]
Mon, 8 Jun 2026 19:46:00 UTC (397 KB)
— Originally published at arxiv.org
Want this in your inbox every morning?
Daily brief at your local 8am — bilingual EN/中文, free.
More from arXiv cs.CL
See more →TriAgent: Divergence-Aware Committees for Cost-Efficient Financial Sentiment Analysis
TriAgent introduces a cost-efficient multi-agent system for financial sentiment analysis, combining VADER, FinBERT, and Qwen2.5. It achieves an F1 score of ~0.87 with significant savings of $9.3M/year at a 10M-user scale compared to GPT-4o-mini, while also detecting hallucinations with an AUC of 0.90.