Abstract
This paper investigates how gender-fair strategies in source texts influence the production of non-binary translations in the Italian to German combination. We use a controlled test set featuring bi nary and non-binary approaches to assess their effectiveness for non-binary render ings in the target language. We also intro duce an automatic evaluation framework that classifies target sentences into four categories: non-binary, binary-gendered, single-gendered, and incoherent. Relying on humananalysis, wecompareReasoning LLMs against standard inference, exam ining whether reasoning improves trans lation quality and automatic evaluation. Our results show that reasoning models are more successful in shifting from binary to non-binary formulations and in handling linguistic challenges such as epicene terms and special characters, although there are no improvements in sentence-level consis tency and evaluation accuracy. A qualita tive analysis of German translations shows that reasoning encourages the reformula tion of source-side strategies through neu tralization, visibility strategies, and para phrasing, resulting in more natural target texts.