Variability in Zero-Shot Cross-Lingual Transfer Robustness Across Model Sizes and Training Regimes

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 robustness of zero-shot cross-lingual transfer performance vary across different model sizes (7B vs. 13B vs. 70B) when trained on English intermediate tasks versus multilingual intermediate tasks, as measured by performance variance across languages in the XTREME-R benchmark?

Autonomous synthesis report generated by Assignee Research. Tribunal consensus score: 9.3/10.