A gradient-attention explainability method produces sequence-level visual and textual saliency maps for free-form answers from large vision-language models, with stronger human-attention alignment and faithfulness than prior baselines.
Visual Explanations via Iterated Integrated Attributions
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
abstract
We introduce Iterated Integrated Attributions (IIA) - a generic method for explaining the predictions of vision models. IIA employs iterative integration across the input image, the internal representations generated by the model, and their gradients, yielding precise and focused explanation maps. We demonstrate the effectiveness of IIA through comprehensive evaluations across various tasks, datasets, and network architectures. Our results showcase that IIA produces accurate explanation maps, outperforming other state-of-the-art explanation techniques.
citation-role summary
baseline 1
citation-polarity summary
fields
cs.CV 1years
2025 1verdicts
CONDITIONAL 1roles
baseline 1polarities
baseline 1representative citing papers
citing papers explorer
-
GLIMPSE: Holistic Cross-Modal Explainability for Large Vision-Language Models
A gradient-attention explainability method produces sequence-level visual and textual saliency maps for free-form answers from large vision-language models, with stronger human-attention alignment and faithfulness than prior baselines.