the open
    research
    Contribute
    Back to the reading lists

    Safety, alignment & interpretability· 2019

    Attention is not Explanation

    Sarthak Jain (first author) · 2019

    Read the original

    arXiv 1902.10186OpenAlex W2934842096

    On the reading lists

    Why it matters

    Challenged the assumption that attention weights faithfully explain neural model decisions.

    Condensed from the expert lists' notes.

    The record

    First released
    2019 (from arXiv 1902.10186)
    Citations
    2,241 citations, as counted by Semantic Scholar logoSemantic Scholar (non-commercial licence)
    Named by
    both expert lists
    Score
    0.424 · computed rank 331 · consensus 0.62, must-read 0, citations percentile 0.681 (2010s · safety), awards 0, existing lists 0. Method

    Data sources

    Catalogue data from OpenAlex (CC0), Crossref (CC0 metadata; no abstracts are shown), arXiv, and Papers With Code (CC BY-SA 4.0), and Semantic Scholar logoSemantic Scholar (used under its non-commercial API licence). Each count names its source. See the method.