| Benchmark | Model | Rank | Results |
|---|---|---|---|
| d4rl-on-d4rl | Reformer | #7 | Average Reward: 63.9 |
| image-generation-on-imagenet-64x64 | Reformer (12 layers) | #44 | Bits per dim: 3.710 |
| image-generation-on-imagenet-64x64 | Reformer (6 layers) | #46 | Bits per dim: 3.740 |
| language-modelling-on-wikitext-103 | Reformer 125M | #60 | Test perplexity: 26.0 |
| open-domain-question-answering-on-searchqa | Locality-Sensitive Hashing | #1 | EM: 66.0 |
| question-answering-on-natural-questions-long | Locality-Sensitive Hashing | #2 | F1: 75.5 |
| question-answering-on-quasart-t | Locality-Sensitive Hashing | #1 | EM: 53.2 |