@inproceedings{chiang-etal-2020-pretrained,
    title = "{P}retrained Language Model Embryology: {T}he Birth of {ALBERT}",
    author = "Chiang, Cheng-Han  and
      Huang, Sung-Feng  and
      Lee, Hung-yi",
    editor = "Webber, Bonnie  and
      Cohn, Trevor  and
      He, Yulan  and
      Liu, Yang",
    booktitle = "Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP)",
    month = nov,
    year = "2020",
    address = "Online",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/jlcl-multiple-ingestion/2020.emnlp-main.553/",
    doi = "10.18653/v1/2020.emnlp-main.553",
    pages = "6813--6828",
    abstract = "While behaviors of pretrained language models (LMs) have been thoroughly examined, what happened during pretraining is rarely studied. We thus investigate the developmental process from a set of randomly initialized parameters to a totipotent language model, which we refer to as the \textit{embryology} of a pretrained language model. Our results show that ALBERT learns to reconstruct and predict tokens of different parts of speech (POS) in different learning speeds during pretraining. We also find that linguistic knowledge and world knowledge do not generally improve as pretraining proceeds, nor do downstream tasks' performance. These findings suggest that knowledge of a pretrained model varies during pretraining, and having more pretrain steps does not necessarily provide a model with more comprehensive knowledge. We provide source codes and pretrained models to reproduce our results at \url{https://github.com/d223302/albert-embryology}."
}