Robustness of Zero-Shot Cross-Lingual Transfer in XTREME-R Across Language Families

Pre-trained multilingual language encoders, such as multilingual BERT and XLM-R, show great potential for zero-shot cross-lingual transfer. However, these multilingual encoders do not precisely align words and phrases across languages. Especially, learning alignments in the multilingual embedding space usually requires sentence-level or word-level parallel corpora, which are expensive to be obtained for low-resource languages. An alternative is to make the multilingual encoders more robust; when fine-tuning the encoder using downstream task, we train the encoder to tolerate noise in the contex

Research goal: How does the robustness of zero-shot cross-lingual transfer in XTREME-R vary across different language families (e.g., Indo-European vs. Uralic) when using hybrid batch training versus purely multilingual training, as measured by the F1 score on NER tasks in the XTREME-R benchmark?

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