Expanding Sparse Tuning for Low Memory Usage

Open paper
Benchmark Model Rank Results
monocular-depth-estimation-on-clevr-distanceSNELL-32 ViT-B/16 (VTAB-1k)–Accuracy: 68.6
object-counting-on-clevr-countSNELL-16 ViT-B/16 (VTAB-1k)–Accuracy: 84.4

Rank counts only results with a code link.