Robustness of mT5-large in Code Generation Under Adversarial Attacks with Zero-Shot Cross-Lingual Transfer

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 mT5-large to adversarial examples in code generation tasks differ between zero-shot cross-lingual transfer with and without intermediate-task training on XGLUE benchmarks, as measured by BLEU score drop under targeted attacks?

Autonomous synthesis report generated by Assignee Research. Tribunal consensus score: 8.2/10.