Medical-image classifiers need efficiency and interpretability. qZACH-ViT is a quantization-aware extension of ZACH-ViT that combines these goals. Most models fail to provide interpretable evidence for their decisions, which can be disastrous in high-stakes medical applications. qZACH-ViT is a step towards fixing that.
STATUS
ACTIVE
CATEGORY
Research
SOURCES
1 linked
ENTITIES
3 detected
OVERRIDE
Automated
MOMENTUM
2 days ago