Mathematical Reasoning on MATH (Score and Strategy Count)
0.863ScorerSIM (Plug-in)
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| rSIM (Plug-in)Reasoner=Llama3.3-70B, Planner=3B plug-in2025.12 | 0.863 | 4 | |
| PromptingReasoner=Llama3.3-70B, Planner=70B2025.12 | 0.84 | 6 | |
| rSIM (Transfer)Reasoner=Llama3.3-70B, Planner=Qwen2.5-1.5B2025.12 | 0.837 | 4 | |
| rSIM (Plug-in)Reasoner=Llama3.3-70B, Planner=1B plug-in2025.12 | 0.832 | 3 | |
| PromptingReasoner=Llama3.3-70B, Planner=3B2025.12 | 0.791 | 7 | |
| PS+Reasoner=Llama3.3-70B, Planner=No2025.12 | 0.783 | 0 | |
| ZeroCoTReasoner=Llama3.3-70B, Planner=No2025.12 | 0.77 | 0 | |
| rSIMReasoner=Llama3.2-1B, Planner=3B2025.12 | 0.615 | 4 | |
| rSIMReasoner=Llama3.2-1B, Planner=Qwen2.5-1.5B2025.12 | 0.591 | 4 | |
| rSIMReasoner=Llama3.2-1B, Planner=1B2025.12 | 0.57 | 3 | |
| PromptingReasoner=Llama3.2-3B, Planner=70B2025.12 | 0.555 | 7 | |
| ZeroCoTReasoner=Llama3.2-3B, Planner=No2025.12 | 0.48 | 0 | |
| PS+Reasoner=Llama3.2-3B, Planner=No2025.12 | 0.475 | 0 | |
| PromptingReasoner=Llama3.2-3B, Planner=3B2025.12 | 0.464 | 7 | |
| PromptingReasoner=Llama3.2-1B, Planner=70B2025.12 | 0.333 | 5 | |
| ZeroCoTReasoner=Llama3.2-1B, Planner=No2025.12 | 0.306 | 0 | |
| PS+Reasoner=Llama3.2-1B, Planner=No2025.12 | 0.282 | 0 | |
| PromptingReasoner=Llama3.2-1B, Planner=3B2025.12 | 0.274 | 7 |