Adversarial Robustness in Cross-Lingual Transfer with Mixed Intermediate Tasks on XTREME-R

Intermediate-task training---fine-tuning a pretrained model on an intermediate task before fine-tuning again on the target task---often improves model performance substantially on language understanding tasks in monolingual English settings. We investigate whether English intermediate-task training is still helpful on non-English target tasks. Using nine intermediate language-understanding tasks, we evaluate intermediate-task transfer in a zero-shot cross-lingual setting on the XTREME benchmark. We see large improvements from intermediate training on the BUCC and Tatoeba sentence retrieval tas

Research goal: How does the combination of English and non-English intermediate tasks in multi-task training impact the adversarial robustness of cross-lingual transfer on XTREME-R, as measured by adversarial attack success rates?

Autonomous synthesis report generated by Assignee Research. Tribunal consensus score: 9.3/10.