arXiv Open Access 2025

Optimising ChatGPT for creativity in literary translation: A case study from English into Dutch, Chinese, Catalan and Spanish

Shuxiang Du Ana Guerberof Arenas Antonio Toral Kyo Gerrits Josep Marco Borillo
Lihat Sumber

Abstrak

This study examines the variability of Chat-GPT machine translation (MT) outputs across six different configurations in four languages,with a focus on creativity in a literary text. We evaluate GPT translations in different text granularity levels, temperature settings and prompting strategies with a Creativity Score formula. We found that prompting ChatGPT with a minimal instruction yields the best creative translations, with "Translate the following text into [TG] creatively" at the temperature of 1.0 outperforming other configurations and DeepL in Spanish, Dutch, and Chinese. Nonetheless, ChatGPT consistently underperforms compared to human translation (HT).

Topik & Kata Kunci

Penulis (5)

S

Shuxiang Du

A

Ana Guerberof Arenas

A

Antonio Toral

K

Kyo Gerrits

J

Josep Marco Borillo

Format Sitasi

Du, S., Arenas, A.G., Toral, A., Gerrits, K., Borillo, J.M. (2025). Optimising ChatGPT for creativity in literary translation: A case study from English into Dutch, Chinese, Catalan and Spanish. https://arxiv.org/abs/2504.18221

Akses Cepat

Lihat di Sumber
Informasi Jurnal
Tahun Terbit
2025
Bahasa
en
Sumber Database
arXiv
Akses
Open Access ✓