Task-Relatedness Effects on Zero-Shot Cross-Lingual Logical Inference Performance

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: What is the impact of task-relatedness (e.g., semantic similarity, task difficulty) between English intermediate tasks and XTREME-R logical inference on zero-shot cross-lingual performance, measured by accuracy differences between aligned vs. misaligned task pairs in the target languages?

Autonomous synthesis report generated by Assignee Research. Tribunal consensus score: 7.7/10.