Paper: openreview.net/forum?id=mkq...

💡What can we do to make self-explanations less ambiguous?

-> We propose to automatically adapt explanations to the task by stitching together SE-GNNs with white-box models and combining their explanations.

- Self-explanations can be "unfaithful" by design

- Models encoding different tasks can produce the same self-explanations, limiting the usefulness of explanations

Studying some popular models, we found that:

- The information that self-explanations convey can radically change based on the underlying task to be explained, which is, however, generally unknown

3. ITS ROLE IN OOD GENERALISATION

Domain-Invariant GNNs make predictions over a domain-invariant subgraph to achieve OOD generalisation. We show that unless this subgraph is also *sufficient*, DIGNNs are not domain-invariant.

5/5

2. HOW GNNs AIM TO ACHIEVE IT

We highlight several architectural design choices of Self-Explainable GNNs favoring information leakage from nodes outside the explanation, and propose mitigations.

4/5

We propose rethinking faithfulness from three essential angles:

1. HOW TO COMPUTE IT

Many ways to compute faithfulness exists, but we show:

- they are not interchangeable
- some of them do not have the desired semantics

3/5

Paper: "Reconsidering Faithfulness in Regular, Self-Explainable and Domain Invariant GNNs"
Link: openreview.net/forum?id=kiO...
Poster session: 26 April 10am

2/5