@inproceedings{hu-etal-2024-generative,
    title = "Generative Pretrained Structured Transformers: Unsupervised Syntactic Language Models at Scale",
    author = "Hu, Xiang  and
      Ji, Pengyu  and
      Zhu, Qingyang  and
      Wu, Wei  and
      Tu, Kewei",
    editor = "Ku, Lun-Wei  and
      Martins, Andre  and
      Srikumar, Vivek",
    booktitle = "Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)",
    month = aug,
    year = "2024",
    address = "Bangkok, Thailand",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/fix-sig-urls/2024.acl-long.145/",
    doi = "10.18653/v1/2024.acl-long.145",
    pages = "2640--2657",
    abstract = "A syntactic language model (SLM) incrementally generates a sentence with its syntactic tree in a left-to-right manner.We present Generative Pretrained Structured Transformers (GPST), an unsupervised SLM at scale capable of being pre-trained from scratch on raw texts with high parallelism. GPST circumvents the limitations of previous SLMs such as relying on gold trees and sequential training. It consists of two components, a usual SLM supervised by a uni-directional language modeling loss, and an additional composition model, which induces syntactic parse trees and computes constituent representations, supervised by a bi-directional language modeling loss. We propose a representation surrogate to enable joint parallel training of the two models in a hard-EM fashion.We pre-train GPST on OpenWebText, a corpus with billion tokens, and demonstrate the superiority of GPST over GPT-2 with a comparable size in numerous tasks covering both language understanding and language generation. Meanwhile, GPST also significantly outperforms existing unsupervised SLMs on left-to-right grammar induction, while holding a substantial acceleration on training."
}