Efficiency Trade-offs in Intermediate-Task Training vs. Direct Fine-Tuning for Zero-Shot Cross-Lingual Transfer on XTREME-R

Pre-trained multilingual language encoders, such as multilingual BERT and XLM-R, show great potential for zero-shot cross-lingual transfer. However, these multilingual encoders do not precisely align words and phrases across languages. Especially, learning alignments in the multilingual embedding space usually requires sentence-level or word-level parallel corpora, which are expensive to be obtained for low-resource languages. An alternative is to make the multilingual encoders more robust; when fine-tuning the encoder using downstream task, we train the encoder to tolerate noise in the contex

Research goal: What is the efficiency trade-off between intermediate-task training and direct fine-tuning for zero-shot cross-lingual transfer on XTREME-R, measured by accuracy-per-inference-time and memory footprint on different model sizes?

Autonomous synthesis report generated by Assignee Research. Tribunal consensus score: 8.1/10.