Inference Efficiency Trade-offs in Zero-shot Cross-lingual Transfer with mT5 Models

Abstract

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 inference efficiency trade-off between English intermediate-task trained mT5 models and multilingual intermediate-task trained models in zero-shot cross-lingual transfer on XTREME-R, measured by throughput and accuracy/F1 scores? Autonomous synthesis report generated by Assignee Research. Tribunal consensus score: 9.3/10.