Long Range Arena
Benchmarks
| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Long-range sequence modeling | Long Range Arena (LRA) | Text Accuracy90.53 | 177 | |
| Long-range Sequence Modeling | Long Range Arena (LRA) (test) | Accuracy (Avg)88.21 | 163 | |
| Efficiency analysis | Long-Range Arena (LRA) | Steps per second99.6 | 84 | |
| Long-sequence modeling | Long Range Arena (LRA) v1 (test) | ListOps63.14 | 66 | |
| Sequence Modeling | Long Range Arena (val) | ListOps Accuracy62.75 | 26 | |
| Long-range dependency modeling | Long Range Arena 100 tokens | ListOps Accuracy85.5 | 14 | |
| Path detection | Long Range Arena (LRA) Pathfinder (test) | Accuracy96.7 | 12 | |
| Mathematical logic sequence modeling | Long Range Arena (LRA) ListOps (test) | Accuracy62.8 | 12 | |
| Text Classification | Long Range Arena (test) | Accuracy82.08 | 7 | |
| Long-range dependency modeling | Long Range Arena (LRA) | Text (4K) Acc66.69 | 5 | |
| Long Range Sequence Modeling | Long Range Arena (LRA) (full suite) | ListOps60.45 | 5 |
Showing 11 of 11 rows