Cross-lingual Inference Efficiency in Multilingual Embeddings with Target-Language Development Sets

Multilingual contextual embeddings have demonstrated state-of-the-art performance in zero-shot cross-lingual transfer learning, where multilingual BERT is fine-tuned on one source language and evaluated on a different target language. However, published results for mBERT zero-shot accuracy vary as much as 17 points on the MLDoc classification task across four papers. We show that the standard practice of using English dev accuracy for model selection in the zero-shot setting makes it difficult to obtain reproducible results on the MLDoc and XNLI tasks. English dev accuracy is often uncorrelate

Research goal: What is the effect of using target-language-specific development sets on the inference efficiency (measured in tokens processed per second) of multilingual contextual embeddings in zero-shot cross-lingual evaluation on the PAWS-X dataset, and how does it compare to using English development sets?

Autonomous synthesis report generated by Assignee Research. Tribunal consensus score: 9.0/10.