Performance Estimation on HLE (Humanity's Last Exam) 0.5% subset
5.6MAEScales++
Evaluation results
| Method | Links | ||
|---|---|---|---|
| Scales++2025.10 | 5.6 | 0.6 | |
| Scales-Clustering OnlyClustering only=true2025.10 | 6.3 | 0.7 | |
| Random2025.10 | 6.6 | 0.3 |
| Method | Links | ||
|---|---|---|---|
| Scales++2025.10 | 5.6 | 0.6 | |
| Scales-Clustering OnlyClustering only=true2025.10 | 6.3 | 0.7 | |
| Random2025.10 | 6.6 | 0.3 |