Standard
Benchmarks
| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Negotiation against a gpt-5.4-high-reasoning seller | Standard held-out (test) | Reward0.4081 | 7 | |
| License Plate Recoverability | Standard DS-S | AUC92 | 5 | |
| GPU Scheduling | Standard Single GPU Poisson arrival process (evaluation) | Reward254.3 | 2 |
Showing 3 of 3 rows