@inproceedings{zhang-etal-2024-incremental,
    title = "Incremental pre-training from smaller language models",
    author = "Zhang, Han  and
      Wang, Hui  and
      Xu, Ruifeng",
    editor = "Wong, Kam-Fai  and
      Zhang, Min  and
      Xu, Ruifeng  and
      Li, Jing  and
      Wei, Zhongyu  and
      Gui, Lin  and
      Liang, Bin  and
      Zhao, Runcong",
    booktitle = "Proceedings of the 10th SIGHAN Workshop on Chinese Language Processing (SIGHAN-10)",
    month = aug,
    year = "2024",
    address = "Bangkok, Thailand",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/jlcl-multiple-ingestion/2024.sighan-1.5/",
    pages = "36--44",
    abstract = "Large language models have recently become a new learning paradigm and led to state-of-the-art performance across a range of tasks. As explosive open-source pre-trained models are available, it is worth investigating how to better utilize existing models. We propose a simple yet effective method, Incr-Pretrain, for incrementally pre-training language models from smaller well-trained source models. Different layer-wise transfer strategies were introduced for model augmentation including parameter copying, initial value padding, and model distillation. Experiments on multiple zero-shot learning tasks demonstrate satisfying inference performance upon transferring and promising training efficiency during continuing pre-training. Compared to training from scratch, Incr-Pretrain can save up to half the training time to get a similar testing loss."
}