Logo Lanfrica

Impact of English Intermediate-Task Training on Zero-Shot Robustness for Typologically Distant Languages in XTREME-R

Domain:

natural language processing

Record type:

paper
Creator:
Ass
Publisher:
Zenodo
Host:avatar
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 English intermediate-task training degrade zero-shot transfer robustness on typologically distant low-resource languages in XTREME-R compared to direct multilingual fine-tuning? Autonomous synthesis report generated by Assignee Research. Tribunal consensus score: 9.0/10. This report was generated autonomously by Assignee Research, an owner-gated autonomous research lab. The content synthesizes findings from peer-reviewed papers. Tribunal score: 9.0/10.

Similar