Feature Attribution in Directed Acyclic Graphs Using Edge Intervention

arXiv:2604.15898v1 Announce Type: new Abstract: For around a decade, non-symbolic methods have been the option of choice when explaining complex machine learning (ML) models.

arXiv:2608.12555v1 Announce Type: new Abstract: Predictive explanation methods attribute a model output; they do not, by themselves, attribute an intervention effect on the

arXiv:2605.15353v1 Announce Type: cross Abstract: Inferring the structure of directed acyclic graphs (DAGs) from data is a central challenge in causal discovery, particularly in

arXiv:2607.05563v1 Announce Type: new Abstract: Interpretable explanation methods in Artificial Intelligence aim to uncover the underlying causes and their effects, enabling a

arXiv:2605.22866v1 Announce Type: new Abstract: Compound AI systems route tasks through hierarchies of specialised components. Attribution is dominated by Shapley-based methods