Tool-Integrated Reasoning on Olympiad Bench
52.75Test AccuracySeq-ALP
Evaluation Results
| Method | Links | |
|---|---|---|
| Seq-ALPtemperature=1.0, sampling=average@32, multi-turn=true2026.03 | 52.75 | |
| Token-ALPtemperature=1.0, sampling=average@32, multi-turn=true2026.03 | 51.55 | |
| Seq-Bypasstemperature=1.0, sampling=average@32, multi-turn=true2026.03 | 51.08 | |
| Token-MIStemperature=1.0, sampling=average@32, multi-turn=true2026.03 | 50.65 | |
| GRPOtemperature=1.0, sampling=average@32, multi-turn=true2026.03 | 49.59 | |
| Seq-MIStemperature=1.0, sampling=average@32, multi-turn=true2026.03 | 48.38 |