@inproceedings{irie-2025-positional,
    title = "Why Are Positional Encodings Nonessential for Deep Autoregressive Transformers? A Petroglyph Revisited",
    author = "Irie, Kazuki",
    editor = "Che, Wanxiang  and
      Nabende, Joyce  and
      Shutova, Ekaterina  and
      Pilehvar, Mohammad Taher",
    booktitle = "Findings of the Association for Computational Linguistics: ACL 2025",
    month = jul,
    year = "2025",
    address = "Vienna, Austria",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/corrections-2025-08/2025.findings-acl.30/",
    doi = "10.18653/v1/2025.findings-acl.30",
    pages = "551--559",
    ISBN = "979-8-89176-256-5",
    abstract = "Do autoregressive Transformer language models require explicit positional encodings (PEs)? The answer is `no' provided they have more than one layer{---}they can distinguish sequences with permuted tokens without the need for explicit PEs. This follows from the fact that a cascade of (permutation invariant) set processors can collectively exhibit sequence-sensitive behavior in the autoregressive setting. This property has been known since early efforts (contemporary with GPT-2) adopting the Transformer for language modeling. However, this result does not appear to have been well disseminated, leading to recent rediscoveries. This may be partially due to a sudden growth of the language modeling community after the advent of GPT-2/3, but perhaps also due to the lack of a clear explanation in prior work, despite being commonly understood by practitioners in the past. Here we review the long-forgotten explanation why explicit PEs are nonessential for multi-layer autoregressive Transformers (in contrast, one-layer models require PEs to discern order information of their inputs), as well as the origin of this result, and hope to re-establish it as a common knowledge."
}