Impact of Monolingual Myanmar Fine-Tuning on Zero-Shot XNLI Accuracy

Zero-shot cross-lingual transfer is a central task in multilingual NLP, allowing models trained in languages with more sufficient training resources to generalize to other low-resource languages. Earlier efforts on this task use parallel corpora, bilingual dictionaries, or other annotated alignment data to improve cross-lingual transferability, which are typically expensive to obtain. In this paper, we propose a simple yet effective method, SALT, to improve the zero-shot cross-lingual transfer of the multilingual pretrained language models without the help of such external data. By incorporati

Research goal: To what extent does fine-tuning on monolingual Myanmar data improve XNLI accuracy for zero-shot cross-lingual transfer compared to multilingual pretraining without Myanmar-specific data?

Autonomous synthesis report generated by Assignee Research. Tribunal consensus score: 7.5/10.