Scaling Intermediate-Task Data Size for Zero-Shot Cross-Lingual Transfer in TyDiQA with mT5-Large

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 scaling of intermediate-task training data size (e.g., 1k vs. 100k samples) affect zero-shot cross-lingual transfer performance on TyDiQA when using mT5-large, measured by exact match accuracy?

Autonomous synthesis report generated by Assignee Research. Tribunal consensus score: 8.7/10.