Gradient-based adversarial fine-tuning for XLM-R robustness in zero-shot cross-lingual classification

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: Does incorporating gradient-based adversarial examples during fine-tuning improve XLM-R's robustness in zero-shot cross-lingual text classification tasks (e.g., PAWS-X) compared to baseline fine-tuning, measured by accuracy under adversarial attacks?

Autonomous synthesis report generated by Assignee Research. Tribunal consensus score: 8.1/10.