Which translator should I choose to go from French to English? CUBBITT is slightly better than DeepL and Google Translate

Article URL: https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0297183
Translation of CUBBITT's result (so in the other direction) followed by Bing/Copilot (GPT4).
In this particular case, I prefer the GPT version. CUBBITT's error in the last sentence (“Further studies are needed to evaluate the effectiveness of MTs with full articles and non-French languages.”)

Our study provides an in-depth study of the performance of MTs-DeepL, Google Translate, and CUBBITT in the specific context of translating medical research abstracts from French to English. This focused evaluation contributes to a nuanced understanding of the applicability of these tools in the clinical field. We not only assessed the accuracy of the translations using established parameters, but also looked at the fluency of the translated text. Our study aims to highlight the practical use of TMs for non-English speaking medical researchers.

We found that all three MTs performed similarly when tested with ROUGE, but CUBBITT performed slightly better than the other two using human evaluation. We also found that in terms of stylistic quality, CUBBITT outperforms the original English text.

Although the study was limited to the analysis of abstracts published in general medical journals and did not evaluate the time required for post-editing, we believe that French-speaking researchers may benefit from translating written articles using DeepL, Google Translate or CUBBITT. French in English. Further studies are needed to assess the effectiveness of MTs with complete essays and non-French languages.

—
“Our study provides an in-depth study of the performance of MTs—DeepL, Google Translate, and CUBBITT—in the specific context of translating medical research abstracts from French to English. This focused evaluation contributes to a more nuanced understanding of the applicability of these tools in the medical field. It not only evaluates the accuracy of translations using established metrics. , we also investigated the fluency of the translated text.Our study aims to highlight the practical use of TMs for non-English speaking medical researchers.

We found that all three MTs behaved similarly when tested with ROUGE, but CUBBITT performed slightly better than the other two using human evaluation. We also found that CUBBITT outperforms the original English text in terms of stylistic quality.

Although the study was limited to the analysis of abstracts published in general medical journals and did not assess the time required for post-editing, we believe that French-speaking researchers may benefit from translating written articles using DeepL, Google Translate or CUBBITT. French in English. Further studies are needed to evaluate the effectiveness of MTs with complete essays and non-French languages. »