Mathematical Reasoning on OOD Math Suite (Math500, GSM8K, Olympiad-Bench)
79.6Pass@1 RateBaseline LLM
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| Baseline LLMSystem Category=Standalone2026.03 | 79.6 | 10,164 | 100 | |
| Brier-SLM-VerbSystem Category=Collaborative2026.03 | 78.8 | 9,264 | 57.4 | |
| COREA *System Category=Collaborative2026.03 | 77.9 | 7,978 | 35.5 | |
| RLVR-SLM-ProbeSystem Category=Collaborative2026.03 | 75.9 | 5,874 | 25.2 | |
| SLM-VerbSystem Category=Collaborative2026.03 | 75.7 | 7,300 | 34.9 | |
| Router+RLVR-SLMSystem Category=Collaborative2026.03 | 74.4 | 4,974 | 20.6 | |
| Brier-SLMSystem Category=Standalone2026.03 | 72.7 | 2,322 | 0 | |
| RLVR-SLM-VerbSystem Category=Collaborative2026.03 | 72.6 | 2,927 | 14 | |
| RLVR-SLM-AvgProbSystem Category=Collaborative2026.03 | 72.4 | 2,206 | 4.5 | |
| L1-SLM*System Category=Standalone2026.03 | 72.1 | 2,589 | 0 | |
| RLVR-SLMSystem Category=Standalone2026.03 | 71.8 | 1,928 | 0 | |
| SLMSystem Category=Standalone2026.03 | 60.4 | 2,622 | 0 |