Performance Estimation on HLE (Humanity's Last Exam) 1% subset
3.5MAEScales++
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| Scales++2025.10 | 3.5 | 0.3 | |
| Scales-Clustering OnlyClustering only=true2025.10 | 4.2 | 0.3 | |
| Random2025.10 | 5.3 | 0.5 |
Loading the SOTA2 catalog…
| Method | Links | ||
|---|---|---|---|
| Scales++2025.10 | 3.5 | 0.3 | |
| Scales-Clustering OnlyClustering only=true2025.10 | 4.2 | 0.3 | |
| Random2025.10 | 5.3 | 0.5 |