Semantic Textual Similarity on STS-B (Leakage, Confidence, Pearson evaluation)
2.68Leakage
SPARSE
Updated 4d ago
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
2026.02 | 2.68 | 1.78 | 0.4817 | |
2026.02 | 19.03 | 1.98 | 0.4005 | |
2026.02 | 20.75 | 1.89 | 0.4003 | |
2026.02 | 32.39 | 8.07 | 0.7614 | |
2026.02 | 49.95 | 13.97 | 0.7185 | |
2026.02 | 53.79 | 15.4 | 0.7187 | |
2026.02 | 64.79 | 36.33 | 0.81 | |
2026.02 | 69.15 | 38.44 | 0.8095 | |
2026.02 | 71.82 | 41.76 | 0.8095 | |
2026.02 | 73.1 | 48.63 | 0.8091 | |
2026.02 | 74.98 | 49.28 | 0.8108 | |
2026.02 | 76.15 | 52.36 | 0.8108 | |
2026.02 | 76.42 | 53.98 | 0.8081 | |
2026.02 | 77.4 | 54.02 | 0.8095 | |
2026.02 | 78.94 | 56.84 | 0.8095 | |
2026.02 | 86.75 | 66.57 | 0.8064 |