Cross-lingual Transfer Performance of Multilingual Models in XTREME-R 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 performance of multilingual models fine-tuned on non-English intermediate tasks compare to English-only fine-tuning in zero-shot cross-lingual transfer on XTREME-R tasks, measured in accuracy scores across language pairs?

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