Fine-tuning Impact on Zero-shot XTREME-R Performance in Low-resource Languages

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: Does fine-tuning on English intermediate tasks before the target task improve zero-shot performance on XTREME-R structured prediction tasks for 7B parameter models compared to 13B models when evaluated on low-resource languages?

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