Cross-lingual Transfer Performance via Intermediate-task Training on English and Non-English 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 combining intermediate-task training on both English and a non-English language improve zero-shot cross-lingual transfer performance on XTREME-R compared to training on either language alone, measured by the overall score on the benchmark?

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