We investigate the role of large language models (LLMs) in promoting gender-inclusive language by evaluating their ability to rewrite biased text and generate counterfactual narratives across multiple languages. We introduce a shared task with two subtasks: gender-inclusive rewriting and counterfactual generation. The task covers five languages English, German, Spanish, Tamil, and Kannada reflecting diverse grammatical gender systems and sociocultural contexts. We release curated word-level and sentence-level datasets to support controlled inclusive generation. A total of 50 teams registered for the shared task, and around 8 teams submitted results. Submissions are evaluated using a hybrid framework combining rubric-based automatic scoring with expert human judgment. Finally, we provide an overview of participating systems and discuss key findings and challenges observed across languages.

The present paper focuses on the analysis of automatic subtitles produced with three different systems. We compare the outputs among each other paying attention to the categories of quality derived from audio-visual translation quality research. Besides that, we also consider comprehensibility of the produced subtitles. Additionally, we analyse the automatic evaluation scores to assess the overall quality. Our results show that automatically generated subtitles subtitles remain below human standards in quality and comprehensibility.

This study examines the machine translation of audio descriptions (AD) as an alternative to producing new AD for audiovisual formats in a foreign language. To assess acceptance and comprehensibility among German users, a survey was conducted with blind and visually impaired participants, examining key AD strategies, such as character description and naming, facial expressions and gestures, and spatio-temporal settings. Participants compared machine-translated English AD with original German AD and provided feedback on these aspects. Results showed overall acceptance of the translated AD, although the original was generally preferred. Findings suggest that AD translation is feasible for the German audience, but further studies are needed on machine translation, production costs, as well as larger-scale user studies.

This contribution investigates whether machine-translated subtitles can be easily distinguished from human-translated ones. For this, we run an experiment using two versions of German subtitles for an English television series: (1)produced manually by professional subtitlers, and (2) translated automatically with a Large Language Model (LLM), i.e., GPT4. Our participants were students of translation studies with varying experience in subtitling and the use of machine translation. We asked participants to guess if the subtitles for a selection of video clips had been translated manually or automatically. Apart from analysing whether machine-translated subtitles are distinguishable from human-translated ones, we also seek for indicators of the differences between human and machine translations. Our results show that although it is overall hard to differentiate between human and machine translations, there are some differences. Notably, the more experience the humans have with translation and subtitling, the more able they are to tell apart the two translation variants.