@inproceedings{zhang-etal-2023-dialog,
    title = "Dialog-Post: Multi-Level Self-Supervised Objectives and Hierarchical Model for Dialogue Post-Training",
    author = "Zhang, Zhenyu  and
      Shen, Lei  and
      Zhao, Yuming  and
      Chen, Meng  and
      He, Xiaodong",
    editor = "Rogers, Anna  and
      Boyd-Graber, Jordan  and
      Okazaki, Naoaki",
    booktitle = "Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)",
    month = jul,
    year = "2023",
    address = "Toronto, Canada",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/ingest-emnlp/2023.acl-long.564/",
    doi = "10.18653/v1/2023.acl-long.564",
    pages = "10134--10148",
    abstract = "Dialogue representation and understanding aim to convert conversational inputs into embeddings and fulfill discriminative tasks. Compared with free-form text, dialogue has two important characteristics, hierarchical semantic structure and multi-facet attributes. Therefore, directly applying the pretrained language models (PLMs) might result in unsatisfactory performance. Recently, several work focused on the dialogue-adaptive post-training (DialPost) that further trains PLMs to fit dialogues. To model dialogues more comprehensively, we propose a DialPost method, Dialog-Post, with multi-level self-supervised objectives and a hierarchical model. These objectives leverage dialogue-specific attributes and use self-supervised signals to fully facilitate the representation and understanding of dialogues. The novel model is a hierarchical segment-wise self-attention network, which contains inner-segment and inter-segment self-attention sub-layers followed by an aggregation and updating module. To evaluate the effectiveness of our methods, we first apply two public datasets for the verification of representation ability. Then we conduct experiments on a newly-labelled dataset that is annotated with 4 dialogue understanding tasks. Experimental results show that our method outperforms existing SOTA models and achieves a 3.3{\%} improvement on average."
}