EMNLP 2021main45 citations

SHAPE: Shifted Absolute Position Embedding for Transformers

Shun Kiyono, Sosuke Kobayashi, Jun Suzuki, Kentaro Inui

Abstract

Position representation is crucial for building position-aware representations in Transformers. Existing position representations suffer from a lack of generalization to test data with unseen lengths or high computational cost. We investigate shifted absolute position embedding (SHAPE) to address both issues. The basic idea of SHAPE is to achieve shift invariance, which is a key property of recent successful position representations, by randomly shifting absolute positions during training. We demonstrate that SHAPE is empirically comparable to its counterpart while being simpler and faster.

BibTeX
@inproceedings{kiyono-etal-2021-shape,
    title = "{SHAPE}: {S}hifted Absolute Position Embedding for Transformers",
    author = "Kiyono, Shun  and
      Kobayashi, Sosuke  and
      Suzuki, Jun  and
      Inui, Kentaro",
    editor = "Moens, Marie-Francine  and
      Huang, Xuanjing  and
      Specia, Lucia  and
      Yih, Scott Wen-tau",
    booktitle = "Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing",
    month = nov,
    year = "2021",
    address = "Online and Punta Cana, Dominican Republic",
    publisher = "Association for Computational Linguistics",
    url = "https://aclanthology.org/2021.emnlp-main.266/",
    doi = "10.18653/v1/2021.emnlp-main.266",
    pages = "3309--3321"
}
SHAPE: Shifted Absolute Position Embedding for Transformers · EMNLP 2021