Edge-based circuits in vision transformers can be automatically recovered to explain and steer model computations for classification and adversarial behaviors.
arXiv preprint arXiv:2404.14349 (2024) 2, 3
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
citation-role summary
background 1
citation-polarity summary
years
2026 2verdicts
UNVERDICTED 2roles
background 1polarities
background 1representative citing papers
Circuit-based metrics from Vision Transformer internals provide better label-free proxies for generalization under distribution shift than existing methods like model confidence.
citing papers explorer
-
Seeing Through Circuits: Faithful Mechanistic Interpretability for Vision Transformers
Edge-based circuits in vision transformers can be automatically recovered to explain and steer model computations for classification and adversarial behaviors.
-
Inside-Out: Measuring Generalization in Vision Transformers Through Inner Workings
Circuit-based metrics from Vision Transformer internals provide better label-free proxies for generalization under distribution shift than existing methods like model confidence.