Hourglass

Reported on 3 benchmarks across 2 tasks · 1 paper

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

Medical3 results

Image GenerationonImageNet 64x64
Bits per dim· 2021-10-26
3.44
best: 4.351 (Logsparse (6 layers))
Hierarchical Transformers Are More Efficient Language Models arXiv:2110.13711
Image GenerationonImageNet 32x32
bpd· 2021-10-26
3.74
best: 3.34 (NFDM)
Hierarchical Transformers Are More Efficient Language Models arXiv:2110.13711
Language Modellingonenwik8
Bit per Character (BPC)· 2021-10-26
0.997
best: 1.67 (LSTM (7 layers))
Hierarchical Transformers Are More Efficient Language Models arXiv:2110.13711