Linguistic Typology and Intermediate-Task Training Efficacy in Zero-Shot Cross-Lingual NLI Benchmarks

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 the effectiveness of English intermediate-task training on zero-shot cross-lingual NLI benchmarks (e.g., XNLI) vary depending on the linguistic typological features of the target languages, and can this be measured using F1-score comparisons across language families?

Autonomous synthesis report generated by Assignee Research. Tribunal consensus score: 8.9/10.