external contest-math suite
Benchmarks
| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Mathematical Reasoning | 120-question external contest-math suite (AIME 2024 I, AIME 2024 II, AIME 2025, AIME 2026, HMMT November 2025) AoPS-derived library | Accuracy Change1.88 | 5 |
Showing 1 of 1 rows