Mathematical Reasoning on AIME Average (mean@30, pass@30)
29.5Mean@30FE-SFT & FC-RL
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| FE-SFT & FC-RLBackbone=Youtu-LLM-2B-Base, Training Stage=RL, Framework=WM-AMT2026.06 | 29.5 | 60 | |
| IWM & RLBackbone=Youtu-LLM-2B-Base, Training Stage=RL2026.06 | 28.6 | 57.8 | |
| FE-SFT & RLBackbone=Youtu-LLM-2B-Base, Training Stage=RL, Framework=WM-AMT2026.06 | 28.4 | 55.5 | |
| SFT & RL*Backbone=Youtu-LLM-2B-Base, Training Stage=RL2026.06 | 28 | 52.2 | |
| FE-SFT & RL*Backbone=Youtu-LLM-2B-Base, Training Stage=RL2026.06 | 28 | 51.1 | |
| SFT & RLBackbone=Youtu-LLM-2B-Base, Training Stage=RL, Framework=WM-AMT2026.06 | 27.8 | 56.6 | |
| State-SFT & RL‡Backbone=Youtu-LLM-2B-Base, Training Stage=RL, Framework=WM-AMT2026.06 | 27.7 | 56.7 | |
| State-SFT & RL†Backbone=Youtu-LLM-2B-Base, Training Stage=RL2026.06 | 27.1 | 56.7 |