Mathematical Reasoning on OlymB
53.93AccuracyADv2
Evaluation Results
| Method | Links | |
|---|---|---|
| ADv2Backbone=Qwen3-14B, Indicator Pool Source=Qwen3-8B-based MAS2026.02 | 53.93 | |
| Single AgentBackbone=Qwen3-14B2026.02 | 52.15 | |
| Dynamic-MASBackbone=Qwen3-14B2026.02 | 51.41 |
Loading the SOTA2 catalog…