Intermediate-Task Dataset Size and Zero-Shot Cross-Lingual Accuracy on XTREME-R for Non-English Target Tasks

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 choice of intermediate-task size (e.g., small vs. large labeled datasets) affect the zero-shot cross-lingual accuracy of models on the XTREME-R benchmark when fine-tuned on non-English target tasks?

Autonomous synthesis report generated by Assignee Research. Tribunal consensus score: 9.0/10.