Impact of Intermediate Task Quantity on Zero-Shot Performance in XTREME-R and XTREME

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 varying the number of intermediate English tasks (1 vs. 3 vs. 9) in multi-task fine-tuning impact zero-shot performance on XTREME-R (reasoning) versus XTREME (understanding), and what is the efficiency-accuracy tradeoff?

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