Attention is not Explanation · HackerPapers discussions

Attention is not Explanation — mentioned by D-Machine

Thu, 08 Jan 2026 04:03:51 GMT

Read the third link / review paper, it is not at all the case that all attention is based on QKV projections. Your terms "sensitivity", "visibility", and "important" are too vague and lack any clear mathematical meaning, so IMO add nothing to any understanding. "Important" also seems factually wrong, given these layers are stacked, so later weights and operations can in fact inflate / reverse things. Deriving e.g. feature importances from self-attention layers remains a highly disputed area (e.g. [1] vs [2], for just the tip of the iceberg). You are also assuming that the importance of…

Paper and discussion history. A mention is not an endorsement.

Attention is not Explanation — mentioned by screye

Tue, 08 Dec 2020 19:23:14 GMT

The pair of these papers: (Don't read them in full.) 1.Attention is not explanation (https://arxiv.org/abs/1902.10186) 2.Attention is not not Explanation (https://arxiv.org/abs/1908.04626) Goes to show the complete lack of agreement between researchers in the explainability space. Most popular packages (allen NLP, google LIT, Captum) use saliency based methods (Integrated gradients) or Attention. The community has fundamental disagreements on whether they capture anything equivalent to importance as humans would understand it. An entire community of fairness, ethics and Computational social…

Paper and discussion history. A mention is not an endorsement.

Attention is not Explanation — HN submission

Thu, 28 Feb 2019 15:44:23 GMT

Attention Is Not Explanation

2 points; 0 discussion replies on HN. Paper and source history. A submission is not an endorsement.