Mathematical Reasoning on AIME ’25 (avg@16, pass@16)
41.5Avg@16Dr. MAS
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| Dr. MASBackbone=Qwen3-8B, Agent Setting=Multi-Agent & LLM Non-Sharing2026.02 | 41.5 | 56.7 | |
| Dr. MASBackbone=Qwen3-8B, Agent Setting=Multi-Agent & LLM Sharing2026.02 | 39.4 | 70 | |
| Dr. MASBackbone=Qwen3-4B, Agent Setting=Multi-Agent & LLM Sharing2026.02 | 38.1 | 63.3 | |
| Dr. MASBackbone=Qwen3-4B, Agent Setting=Multi-Agent & LLM Non-Sharing2026.02 | 38.1 | 66.7 | |
| GRPOBackbone=Qwen3-4B, Agent Setting=Multi-Agent & LLM Non-Sharing2026.02 | 35.6 | 63.3 | |
| GRPOBackbone=Qwen3-4B, Agent Setting=Single-Agent2026.02 | 33.1 | 56.7 | |
| GRPOBackbone=Qwen3-8B, Agent Setting=Single-Agent2026.02 | 32.7 | 50 | |
| GRPOBackbone=Qwen3-8B, Agent Setting=Multi-Agent & LLM Non-Sharing2026.02 | 31.8 | 53.3 | |
| GRPOBackbone=Qwen3-4B, Agent Setting=Multi-Agent & LLM Sharing2026.02 | 31.4 | 53.3 | |
| GRPOBackbone=Qwen3-8B, Agent Setting=Multi-Agent & LLM Sharing2026.02 | 31.4 | 53.3 |