echocardiography 2concept bottleneck models 1deep video models 1ejection fraction 1interpretability 1model attribution 1temporal localization 1transformer encoder 1volume estimation 1
From the 2 of 3 linked papers with an AI index.
3 papers
cs.AI2026
Loss Invariance Determines What Concept Layers Encode: Volume Grounding in Echocardiography
Hyunkyung Han, Min Jung Kim
The paper investigates how loss invariance affects the encoding of volume concepts in a transformer-based echocardiography model, showing that without explicit volume supervision t…
cs.CV2026
Ablation-Corrected Evaluation of Attribution Maps in Echocardiographic Ejection-Fraction Models
Hyunkyung Han, Min Jung Kim
The paper audits attribution methods for deep video models that estimate left‑ventricular ejection fraction from echocardiography, showing that while the models correctly focus on…
cs.CV2025
Rethinking Random Masking in Self-Distillation on ViT
Jihyeon Seong, Hyunkyung Han
Vision Transformers (ViTs) have demonstrated remarkable performance across a wide range of vision tasks. In particular, self-distillation frameworks such as DINO have contributed s…