This video is a group presentation for the Explainable and Trustworthy AI module, presenting the paper “Attention Is Not Explanation.” We present the key ideas of the paper, explain why attention mechanisms should not be considered reliable explanations, and discuss their implications for trustworthy AI.
Presented by:
Tom Louis Klein, Szymon Stano and Yasmine Mahdoui
Infos
- Yasmine Mahdoui
- 11 janvier 2026 17:57
- Autres
- Anglais
Commentaire(s)