Mathematical Reasoning on MATH (test) (Brier Score + AUROC)
0.051Brier ScoreMSV
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| MSVBackbone=Llama-3.2-1B-Instruct, Verifier Type=Multi-Sequence Verifier, Number of Sequences (N)=642026.03 | 0.051 | 0.9513 | |
| MSVBackbone=Llama-3.2-1B-Instruct, Verifier Type=Multi-Sequence Verifier, Number of Sequences (N)=162026.03 | 0.0596 | 0.9222 | |
| MSVBackbone=Llama-3.2-1B-Instruct, Verifier Type=Multi-Sequence Verifier, Number of Sequences (N)=12026.03 | 0.1192 | 0.7654 | |
| ProbeBackbone=Llama-3.2-1B-Instruct, Verifier Type=Probe2026.03 | 0.1236 | 0.7856 |