S2
Benchmarks
| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Clustering | S2 N(150) | ACC100 | 20 | |
| Adversarial Attack | S2 ZH interview | ASR20.8 | 15 | |
| Adversarial Attack | S2 EN interview | Attack Success Rate (ASR)84 | 15 | |
| Cluster number prediction | S2 | Cluster Count (NC)23 | 15 | |
| Synthetic Function Optimization | S2 Contig. Rosen. | Median LogGap1.8876 | 14 | |
| Clustering | S2-L N(150) | Accuracy100 | 10 | |
| Reward prediction | S2 10-task benchmark unconventional | Demo L MSE0.031 | 7 | |
| Cluster count selection | S2 | Selected K Value6 | 7 | |
| Flood Segmentation | S2 multispectral | AUPRC95.57 | 6 | |
| Nuclear Segmentation | S2 | Coverage94.26 | 6 | |
| Point-level consensus correctness prediction | S2 | AUPRC93.1 | 4 | |
| Code Search | S2 | ANR34.24 | 3 | |
| Intrinsic-geometry transport | S2 | Plan KL0.05 | 3 | |
| Task performance and governance | S2 LowInfo | CDL45.3 | 2 | |
| Cluster count selection | S2 | S_mM Score6 | 2 |
Showing 15 of 15 rows