Tool Usage on Meta-Tool Single
77.7AccuracyPhi-4-Reasoning
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| Phi-4-ReasoningBase Model=Phi-4-Reasoning, Stopping Strategy=Base2025.10 | 77.7 | — | |
| THINKBRAKEBase Model=Phi-4-Reasoning, Stopping Strategy=THINKBRAKE2025.10 | 77.2 | -2.5 | |
| NoWaitBase Model=Phi-4-Reasoning, Stopping Strategy=NoWait2025.10 | 76.6 | -8.5 | |
| Qwen3-4BBase Model=Qwen3-4B, Stopping Strategy=Base2025.10 | 74.3 | — | |
| THINKBRAKEBase Model=Qwen3-4B, Stopping Strategy=THINKBRAKE2025.10 | 72.7 | -1.4 | |
| NoWaitBase Model=Qwen3-4B, Stopping Strategy=NoWait2025.10 | 72.5 | -3.1 | |
| THINKBRAKEBase Model=Qwen3-4B-Thinking, Stopping Strategy=THINKBRAKE2025.10 | 72.4 | -20.6 | |
| THINKBRAKE-pBase Model=Qwen3-4B-Thinking, Stopping Strategy=THINKBRAKE-p2025.10 | 70.8 | -25.8 | |
| Qwen3-4B-ThinkingBase Model=Qwen3-4B-Thinking, Stopping Strategy=Base2025.10 | 69.7 | — | |
| NoWaitBase Model=Qwen3-4B-Thinking, Stopping Strategy=NoWait2025.10 | 68.9 | 0.1 | |
| ThinkLessBase Model=Qwen3-4B-Thinking, Stopping Strategy=ThinkLess2025.10 | 67.4 | -100 | |
| THINKBRAKEBase Model=Qwen3-14B, Stopping Strategy=THINKBRAKE2025.10 | 66.3 | -1.6 | |
| NoWaitBase Model=DeepSeek-R1-7B, Stopping Strategy=NoWait2025.10 | 64.9 | 0.6 | |
| Qwen3-32BBase Model=Qwen3-32B, Stopping Strategy=Base2025.10 | 64.8 | — | |
| THINKBRAKEBase Model=Qwen3-32B, Stopping Strategy=THINKBRAKE2025.10 | 64.4 | -1.5 | |
| NoWaitBase Model=Qwen3-32B, Stopping Strategy=NoWait2025.10 | 63.8 | 3.6 | |
| THINKBRAKEBase Model=DeepSeek-R1-7B, Stopping Strategy=THINKBRAKE2025.10 | 63.4 | -7.4 | |
| Qwen3-14BBase Model=Qwen3-14B, Stopping Strategy=Base2025.10 | 63.3 | — | |
| DeepSeek-R1-7BBase Model=DeepSeek-R1-7B, Stopping Strategy=Base2025.10 | 63.2 | — | |
| NoWaitBase Model=Qwen3-14B, Stopping Strategy=NoWait2025.10 | 61.2 | -2.4 |