feature interactions

1 paper across 1 session

Poster Session 6

Friday, December 5, 2025 · 4:30 PM → 7:30 PM

Explaining Similarity in Vision-Language Encoders with Weighted Banzhaf Interactions

#1112 · Hubert Baniecki, Maximilian Muschalik, Fabian Fumagalli, Barbara Hammer, Eyke Hüllermeier, Przemyslaw Biecek

We introduce faithful interaction explanations of CLIP and SigLIP models (FIxLIP), offering a unique perspective on interpreting image–text similarity predictions.