Success Rate Prediction on MATH
83.3Perturbed SRTRUE (DAG-based predictor)
Evaluation Results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| TRUE (DAG-based predictor)Category=Number Theory2026.02 | 83.3 | 75 | 0.75 | — | 0.64 | |
| Baseline CoT predictorCategory=Number Theory2026.02 | 83.3 | 75 | — | 1.39 | — | |
| TRUE (DAG-based predictor)Category=Prealgebra2026.02 | 50 | 87.5 | 1.05 | — | 10.46 | |
| Baseline CoT predictorCategory=Prealgebra2026.02 | 50 | 87.5 | — | 11.51 | — | |
| TRUE (DAG-based predictor)Category=Counting & Prob.2026.02 | 50 | 75 | 1.72 | — | 6.97 | |
| Baseline CoT predictorCategory=Counting & Prob.2026.02 | 50 | 75 | — | 8.69 | — | |
| TRUE (DAG-based predictor)Category=Average2026.02 | 46.7 | 65 | 0.85 | — | 3.78 | |
| Baseline CoT predictorCategory=Average2026.02 | 46.7 | 65 | — | 4.63 | — | |
| TRUE (DAG-based predictor)Category=Intermediate Alg.2026.02 | 33.3 | 50 | 0.3 | — | 0.45 | |
| Baseline CoT predictorCategory=Intermediate Alg.2026.02 | 33.3 | 50 | — | 0.75 | — | |
| TRUE (DAG-based predictor)Category=Precalculus2026.02 | 16.7 | 37.5 | 0.43 | — | 0.36 | |
| Baseline CoT predictorCategory=Precalculus2026.02 | 16.7 | 37.5 | — | 0.79 | — |