Batch Size Impact on Zero-Shot Cross-Lingual Transfer Trade-Offs in XTREME-R
Description
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 adjusting the batch size during intermediate-task training affect inference latency and F1 score trade-offs for zero-shot cross-lingual transfer on the XTREME-R benchmark?
Autonomous synthesis report generated by Assignee Research. Tribunal consensus score: 7.5/10.
Notes
Files
paper.pdf
Files
(87.1 kB)
| Name | Size | Download all |
|---|---|---|
|
md5:011dbd657c6424dbd04981451eff8251
|
87.1 kB | Preview Download |