Sequential vs. Single Intermediate-Task Training for XLM-R-Base Zero-Shot Accuracy in 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: What is the impact of sequential intermediate-task training (vs. single intermediate-task training) on XLM-R-Base's zero-shot accuracy in XTREME while maintaining inference efficiency within 5% of the baseline?

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