SC
Benchmarks
| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Narrative Understanding | SC | First-Token Accuracy98.7 | 24 | |
| Integer Linear Programming | SC | Feasibility Rate (FR)100 | 13 | |
| Set Covering | SC-4000 | Objective170.3 | 10 | |
| Set Covering | SC-2000 | Objective Value291.5 | 10 | |
| Scientific Claims | SC | R Score87.3 | 10 | |
| Temporal scRNA-seq data modeling | SC Hard Schiebinger | Performance at t=513.66 | 9 | |
| scRNA-seq Extrapolation | SC Schier (Medium (Extrapolation)) | Performance at t=1615.24 | 9 | |
| scRNA-seq Interpolation | SC Easy Interpolation | Error at t=510.19 | 9 | |
| Multi-objective shortest path | SC 100 | Hypervolume (HV)0.7 | 9 | |
| Sentence Classification | SC | MNLI Accuracy90.53 | 9 | |
| Resource Constrained Shortest Path | SC 100 | Objective Value7.77 | 8 | |
| Set Covering | SC Large Scale | Gap Improvement400 | 8 | |
| Set Covering | SC Medium Scale | Gap Improvement200 | 8 | |
| Set Cover | SC 2000 elements (Out-of-Distribution) | Objective Value9.3 | 7 | |
| Set Cover | SC In-Distribution 2000 elements | Objective Value291.5 | 7 | |
| Regression | SC | Average Training Time (s)2.92 | 6 | |
| Speech Classification | SC10 Raw (test) | Accuracy98.32 | 6 | |
| Unconditional generation of event sequences | SC synthetic (test) | MMD5.8 | 5 | |
| Temporal Point Process Generation | SC (test) | MMD5.8 | 5 | |
| Adversarial NSFW Image Generation | SC (test) | ASR-2570 | 5 | |
| Density Estimation | SC (test) | MMD0.08 | 5 | |
| Density estimation | SC Synthetic (test) | Wasserstein Distance0 | 5 | |
| Speech Classification | SC10 Raw 0.5x (test) | Accuracy96.3 | 5 | |
| Sarcasm Detection | SC GEN v2 | F1 Score75 | 5 | |
| Sarcasm Detection | SC v1 | F1 Score69 | 5 |
Showing 25 of 29 rows