ART

Reported on 7 benchmarks across 2 tasks · 1 paper · 2 SOTA

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Robots5 results

Vision and Language NavigationonVLN Challenge
error
4.09
best: 1.61 (human)
Vision and Language NavigationonVLN Challenge
length
12.35
best: 1257.38 (Speaker-Follower)
Vision and Language NavigationonVLN Challenge
oracle success
0.7
best: 1 (FOAM-Beam Search)
Vision and Language NavigationonVLN Challenge
spl
0.57
best: 0.76 (human)
Vision and Language NavigationonVLN Challenge
success
0.63
best: 0.86 (human)

Computer Vision2 results

Object Localizationon CUB-200-2011
Top-1 Error Rate· 2019-11-29
34.8
SOTA
Attributional Robustness Training using Input-Gradient Spatial Alignment arXiv:1911.13073
Object Localizationon CUB-200-2011
Top-1 Localization Accuracy· 2019-11-29
65.22
best: 87 (GenPromp)
SOTA
Attributional Robustness Training using Input-Gradient Spatial Alignment arXiv:1911.13073