Optimising ChatGPT for creativity in literary translation: A case study from English into Dutch, Chinese, Catalan and Spanish
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866912346503380992 |
|---|---|
| author | Du, Shuxiang Arenas, Ana Guerberof Toral, Antonio Gerrits, Kyo Borillo, Josep Marco |
| author_facet | Du, Shuxiang Arenas, Ana Guerberof Toral, Antonio Gerrits, Kyo Borillo, Josep Marco |
| contents | 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). |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2504_18221 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Optimising ChatGPT for creativity in literary translation: A case study from English into Dutch, Chinese, Catalan and Spanish Du, Shuxiang Arenas, Ana Guerberof Toral, Antonio Gerrits, Kyo Borillo, Josep Marco Computation and Language 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). |
| title | Optimising ChatGPT for creativity in literary translation: A case study from English into Dutch, Chinese, Catalan and Spanish |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2504.18221 |