Cross-lingual Zero-shot Accuracy in 175B Models with Varying Intermediate Task Data Sizes

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 intermediate-task training data size (10K vs. 100K vs. 1M examples) affect the zero-shot cross-lingual accuracy of 175B parameter models on the XTREME-R benchmark, compared to the original XTREME benchmark?

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