iiinova — where your digital self keeps evolving

Steve Azzolin

@steveazzolin.bsky.social

209 followers 127 following 14 posts

ELLIS PhD student @ UNITN/UniCambridge || Prev. Visiting Research Student at UniCambridge || Prev. Research intern at SISLab

Posts Replies Media

Steve Azzolin @steveazzolin.bsky.social · 13/07/2025

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

openreview.net

Beyond Topological Self-Explainable GNNs: A Formal Explainability...

Self-Explainable Graph Neural Networks (SE-GNNs) are popular explainable-by-design GNNs, but their explanations' properties and limitations are not well understood. Our first contribution fills...

000

Steve Azzolin @steveazzolin.bsky.social · 13/07/2025

💡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.

100

Steve Azzolin @steveazzolin.bsky.social · 13/07/2025

- Self-explanations can be "unfaithful" by design

100

Steve Azzolin @steveazzolin.bsky.social · 13/07/2025

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

100

Steve Azzolin @steveazzolin.bsky.social · 13/07/2025

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

100

Steve Azzolin @steveazzolin.bsky.social · 13/07/2025

🤔 What are the properties of self-explanations in GNNs? What can we expect from them? We investigate this in our #ICML25 paper. Come to have a chat at poster session 5, Thu 17 11 am. w. Sagar Malhotra @andreapasspr.bsky.social @looselycorrect.bsky.social

110

Steve Azzolin @steveazzolin.bsky.social · 25/04/2025

Happening tomorrow! Poster number 508 Saturday's session 10-12:30

001

Steve Azzolin @steveazzolin.bsky.social · 17/04/2025

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

000

Steve Azzolin @steveazzolin.bsky.social · 17/04/2025

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

100

Steve Azzolin @steveazzolin.bsky.social · 17/04/2025

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

100

Steve Azzolin @steveazzolin.bsky.social · 17/04/2025

Paper: "Reconsidering Faithfulness in Regular, Self-Explainable and Domain Invariant GNNs" Link: openreview.net/forum?id=kiO... Poster session: 26 April 10am 2/5

100

Steve Azzolin @steveazzolin.bsky.social · 17/04/2025

Faithfulness of GNN explanations isn’t one-size-fits-all🧢 Our last @iclr-conf.bsky.social paper breaks it down across: 1. Evaluation metrics 2. Model implementations 3. OOD generalisation w: Antonio L. @looselycorrect.bsky.social @andreapasserini.bsky.social 1/5

163

Steve Azzolin @steveazzolin.bsky.social · 08/12/2024

Kudos to the organisers for setting up the poster session in the fanciest room I've ever seen👀

130

Steve Azzolin @steveazzolin.bsky.social · 02/12/2024

Hello World!

170