@inproceedings{fu-etal-2022-adapterbias,
    title = "{A}dapter{B}ias: Parameter-efficient Token-dependent Representation Shift for Adapters in {NLP} Tasks",
    author = "Fu, Chin-Lun  and
      Chen, Zih-Ching  and
      Lee, Yun-Ru  and
      Lee, Hung-yi",
    editor = "Carpuat, Marine  and
      de Marneffe, Marie-Catherine  and
      Meza Ruiz, Ivan Vladimir",
    booktitle = "Findings of the Association for Computational Linguistics: NAACL 2022",
    month = jul,
    year = "2022",
    address = "Seattle, United States",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/jlcl-multiple-ingestion/2022.findings-naacl.199/",
    doi = "10.18653/v1/2022.findings-naacl.199",
    pages = "2608--2621",
    abstract = "Transformer-based pre-trained models with millions of parameters require large storage. Recent approaches tackle this shortcoming by training adapters, but these approaches still require a relatively large number of parameters. In this study, AdapterBias, a surprisingly simple yet effective adapter architecture, is proposed. AdapterBias adds a token-dependent shift to the hidden output of transformer layers to adapt to downstream tasks with only a vector and a linear layer. Extensive experiments are conducted to demonstrate the effectiveness of AdapterBias. The experiments show that our proposed method can dramatically reduce the trainable parameters compared to the previous works with a minimal decrease in task performances compared with fine-tuned pre-trained models. We further find that AdapterBias automatically learns to assign more significant representation shifts to the tokens related to the task in consideration."
}