| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Diagnosis Prediction | Held-out (test) | Median AUROC0.932 | 66 | |
| Sycophancy Mitigation | Held-out | BRR43.6 | 40 | |
| Multi-path Speculative Decoding | Held-out (test) | Average Block Efficiency6.84 | 24 | |
| Tissue Box | held-out N=50 randomized configurations (test) | Hand RMS Jerk0.122 | 22 | |
| Toy Rearrangement | held-out N=50 randomized configurations (test) | Hand RMS Jerk2.94 | 22 | |
| Peach Preparation | held-out N=50 randomized configurations (test) | Hand RMS Jerk4.086 | 22 | |
| Apple Preparation | held-out N=50 randomized configurations (test) | Hand RMS Jerk8.725 | 22 | |
| Bargaining | Held-Out (test) | Reward0.7664 | 16 | |
| Query routing and tool-calling accuracy evaluation | Held-out 12,282 examples (test) | Accuracy89.39 | 15 | |
| Adversarial Attack Detection | Held-out n = 560 (20% stratified test) | Detection Rate89.1 | 11 | |
| Steak Season. | Held-out (test) | RMS Jerk2.132 | 11 | |
| Response Quality Evaluation | Held-out n=300 (test) | Pearson Correlation Coefficient0.747 | 8 | |
| Win probability forecasting | Held-out 2024 n=5185 (held-out) | Brier Score0.1355 | 7 | |
| Inverse Kinematics | Held-out (test) | Mean Error (mm)0 | 7 | |
| SUVR-to-PET Retrieval | Held-out 153 subjects (test) | Rank-1 Retrieval92.2 | 6 | |
| PET-to-SUVR Retrieval | Held-out 153 subjects (test) | R@177.8 | 6 | |
| SUVR Prediction | Held-out 153 subjects (test) | MAE0.07 | 6 | |
| Tone Mapping | Held-out (test) | PSNR40.59 | 6 | |
| Clinical case generation | Held-out (test) | BLEU-418.98 | 6 | |
| Actuarial projection | Held-out (test) | MAE0.0198 | 5 | |
| Selective Classification | Held-out (test) | Coverage100 | 5 | |
| Pairwise preference ranking | Held-out | ELO Score1,187 | 5 | |
| License Plate Recognition | held-out (test) | Plate Accuracy92.3 | 5 | |
| Event-level market-impact prediction | Held-out 2021-2023 (test) | Non-neutral F135.6 | 4 | |
| Binary-level classification | held-out (test) | Accuracy98.4 | 4 |