notesum.ai
Published at May 10Language Model as Visual Explainer
NeurIPS
Released Date: May 10, 2024
Authors: Xingyi Yang1, Xinchao Wang1
Aff.: 1National University of Singapore
Arxiv: https://openreview.net/pdf/74362df179f20f43d867bd8fcfb34a4d06911c90.pdf

| Network | CIFAR-10 | CIFAR-100 | ImageNet | ||||||
|---|---|---|---|---|---|---|---|---|---|
| TrDec | SubTree | LVX | TrDec | SubTree | LVX | TrDec | SubTree | LVX | |
| RN-18 | -0.224 | -0.393 | -0.971 | -0.246 | -0.446 | -0.574 | -0.298 | -0.548 | -0.730 |
| RN-50 | -0.236 | -0.430 | -1.329 | -0.256 | -0.500 | -1.170 | -0.317 | -0.588 | -1.186 |
| ViT-S 16 | -0.244 | -0.467 | -1.677 | -0.266 | -0.527 | -1.073 | -0.330 | -0.626 | -1.792 |