Zero-shot Cross-lingual Performance of XLM-R Base with Intermediate Task Fine-tuning

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 zero-shot cross-lingual performance of XLM-R Base change when using intermediate-task fine-tuning on code-related tasks (e.g., CodeXGLUE) versus NLP tasks (e.g., Paws-X) for downstream XTREME-R classification tasks?

Autonomous synthesis report generated by Assignee Research. Tribunal consensus score: 9.2/10.