Semantic Textual Similarity on STS Task Family Aggregated (test)
0.7Spearman's RhoFLARE
Evaluation results
| Method | Links | ||
|---|---|---|---|
| FLARE2026.04 | 0.7 | 0.7 | |
| Silhouette2026.04 | 0.56 | 0.53 | |
| Uniformity2026.04 | 0.01 | 0.7 | |
| EMIR2026.04 | -0.06 | -0.08 | |
| IsoScore2026.04 | -0.33 | -0.14 |
| Method | Links | ||
|---|---|---|---|
| FLARE2026.04 | 0.7 | 0.7 | |
| Silhouette2026.04 | 0.56 | 0.53 | |
| Uniformity2026.04 | 0.01 | 0.7 | |
| EMIR2026.04 | -0.06 | -0.08 | |
| IsoScore2026.04 | -0.33 | -0.14 |