Impact of XLM-R Model Scaling with Adversarial Pre-training on Zero-shot Cross-lingual Performance in XTREME-R

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: What is the impact of scaling the model size of XLM-R with adversarial pre-training on zero-shot cross-lingual performance in XTREME-R, measured by BLEU and F1 scores across high-, mid-, and low-resource languages?

Autonomous synthesis report generated by Assignee Research. Tribunal consensus score: 8.5/10.