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 reaResearch goal: Does the effectiveness of WEAM's alignment mechanism scale with the number of languages included in pre-training, and what is the trade-off between model size and zero-shot transfer performance on XTREME-R?Autonomous synthesis report generated by Assignee Research. Tribunal consensus score: 7.5/10.