Performance Estimation on HLE (Humanity's Last Exam) 2% subset
2.9MAEScales++
Evaluation results
| Method | Links | ||
|---|---|---|---|
| Scales++2025.10 | 2.9 | 0.3 | |
| Random2025.10 | 3.2 | 0.4 | |
| Scales-Clustering OnlyClustering only=true2025.10 | 3.4 | 0.2 |
| Method | Links | ||
|---|---|---|---|
| Scales++2025.10 | 2.9 | 0.3 | |
| Random2025.10 | 3.2 | 0.4 | |
| Scales-Clustering OnlyClustering only=true2025.10 | 3.4 | 0.2 |