Mathematical Reasoning on OlymMATH Hard (accuracy)
26.25Accuracy (OlymMATH Hard)Fixed-MAS + Self-Refine
Evaluation Results
| Method | Links | |
|---|---|---|
| Fixed-MAS + Self-RefineFramework=Fixed-MAS, Technique=Self-Refine2026.02 | 26.25 | |
| Fixed-MAS + ADv2Framework=Fixed-MAS, Technique=ADv22026.02 | 23.75 | |
| Fixed-MAS + ADv1Framework=Fixed-MAS, Technique=ADv12026.02 | 20 | |
| Dynamic-MAS + Multi-TAGFramework=Dynamic-MAS, Technique=Multi-TAG2026.02 | 18 | |
| Fixed-MASFramework=Fixed-MAS2026.02 | 17.5 | |
| Fixed-MAS + Multi-TAGFramework=Fixed-MAS, Technique=Multi-TAG2026.02 | 17.5 | |
| Dynamic-MAS + ADv2Framework=Dynamic-MAS, Technique=ADv22026.02 | 17 | |
| Fixed-MAS + PRMFramework=Fixed-MAS, Technique=PRM2026.02 | 16.25 | |
| Single AgentFramework=Single Agent2026.02 | 16 | |
| Dynamic-MASFramework=Dynamic-MAS2026.02 | 16 | |
| Single Agent + CoTFramework=Single Agent, Technique=Chain-of-Thought2026.02 | 15 | |
| Dynamic-MAS + Self-RefineFramework=Dynamic-MAS, Technique=Self-Refine2026.02 | 15 | |
| Dynamic-MAS + PRMFramework=Dynamic-MAS, Technique=PRM2026.02 | 15 |