Sparse, Efficient and Explainable Data Attribution with DualXDA
Galip Ümit Yolcu, Moritz Weckbecker, Thomas Wiegand, Wojciech Samek, Sebastian Lapuschkin
Action editor: Steffen Schneider
https://openreview.net/forum?id=qfx81N884A
#attributions #attribution #dualda
Explaining Bayesian Neural Networks
Kirill Bykov, Marina MC Höhne, Adelaida Creosteanu et al.
Action editor: Quanshi Zhang
https://openreview.net/forum?id=ZxsR4t3wJd
#explanations #attributions #ai
Les #enseignants abandonnés par l' #institution.
Présence permanente pour les #exploiter et leur #demander des #services, de préférence :
* #gratuits (au nom d'une bonne mentalité)
* en dehors de leurs #attributions,
Mais absence totale quand ils s'agit de les épauler et les soutenir.
A Dual-Perspective Approach to Evaluating Feature Attribution Methods
Yawei Li, Yang Zhang, Kenji Kawaguchi, Ashkan Khakzar, Bernd Bischl, Mina Rezaei
Action editor: Dumitru Erhan
https://openreview.net/forum?id=znlTP5RLur
#attributions #attribution #feature
Participants rated GPT-4's moral reasoning as superior in quality to humans' across dimensions like virtuousness, intelligence, & trustworthiness.
#AI #attributions #biases #MoralReasoning #TuringTest
#Research: Humans are already attributing #consciousness to #LLMs (like ChatGPT).
Bad news: it's not correct.
Good news: Humans may be open to synthetic consciousness
#AI #perceptions #attributions #trust #relationships
www.ethicalpsychology.com/2023/12/folk...
#research - Cherished values increase the stringency of moral character attributions
If you violate an important personal or cultural #value , people are more likely to judge you with poor moral #character
#attributions #biases #DecisionMaking #judgments
www.ethicalpsychology.com/2023/11/one-...
