Mathematical Problem Solving on IneqMath (IM)
76Exact Match AccuracyDeepMath (vanilla)
Evaluation Results
| Method | Links | |
|---|---|---|
| DeepMath (vanilla)Optimization=vanilla2026.02 | 76 | |
| DeepMath (GRPO)Optimization=GRPO2026.02 | 73 | |
| DeepMath (RULES)Optimization=RULES2026.02 | 71 | |
| Qwen2.5 (vanilla)Optimization=vanilla2026.02 | 55 | |
| Qwen2.5 (GRPO)Optimization=GRPO2026.02 | 54 | |
| Qwen2.5 (RULES)Optimization=RULES2026.02 | 53 | |
| OLMO2 (vanilla)Optimization=vanilla2026.02 | 29 | |
| OLMO2 (RULES)Optimization=RULES2026.02 | 24 | |
| OLMO2 (GRPO)Optimization=GRPO2026.02 | 23 | |
| Llama-3.1 (vanilla)Optimization=vanilla2026.02 | 22 | |
| Llama-3.1 (GRPO)Optimization=GRPO2026.02 | 18 | |
| Llama-3.1 (RULES)Optimization=RULES2026.02 | 16 |