Do Not Trust Additive Explanations

arxiv.orgsite

from

XAI

added

2026-10-10

likes

0

similar

1. ▲
   Attention is not Explanation arxiv.org

   site category #xai link ↗
   
2. ▲
   Explanation in Artificial Intelligence: Insights from the Social Sciences arxiv.org

   site category #xai link ↗
   
3. ▲
   Simonynan, et. al arxiv.org

   site category #xai link ↗
   
4. ▲
   Please Stop Permuting Features An Explanation and Alternatives arxiv.org

   site category #xai link ↗
   
5. ▲
   Attention is not --not-- Explanation arxiv.org

   site category #xai link ↗
   
6. ▲
   Hara, et. al. arxiv.org

   site category #xai link ↗
   

XAI › Papers > Critiques: “Authors argue that addditive explanations (e.g. LIME, SHAP, Break Down) fail to take feature ineractions into account and are thus unreliable.”