The introduction of pretrained cross-lingual language models brought decisive improvements to multilingual NLP tasks. However, the lack of labelled task data necessitates a variety of methods aiming to close the gap to high-resource languages. Zero-shot methods in particular, often use translated task data as a training signal to bridge the performance gap between the source and target language(s). We introduce XeroAlign, a simple method for task-specific alignment of cross-lingual pretrained transformers such as XLM-R. XeroAlign uses translated task data to encourage the model to generate simResearch goal: How does the alignment of multilingual language models with human feedback impact zero-shot cross-lingual performance in text classification tasks compared to alignment using English-only feedback?Autonomous synthesis report generated by Assignee Research. Tribunal consensus score: 8.2/10.