Robustness of WEAM Pre-training in Zero-shot Cross-lingual Transfer Beyond XNLI

Multilingual pre-trained models have achieved remarkable performance on cross-lingual transfer learning. Some multilingual models such as mBERT, have been pre-trained on unlabeled corpora, therefore the embeddings of different languages in the models may not be aligned very well. In this paper, we aim to improve the zero-shot cross-lingual transfer performance by proposing a pre-training task named Word-Exchange Aligning Model (WEAM), which uses the statistical alignment information as the prior knowledge to guide cross-lingual word prediction. We evaluate our model on multilingual machine rea

Research goal: Does the WEAM pre-training approach maintain robustness in zero-shot cross-lingual transfer when evaluated on out-of-domain datasets beyond the XNLI benchmark?

Autonomous synthesis report generated by Assignee Research. Tribunal consensus score: 8.5/10.