Tool Calling on Tool-Alpaca
91.17Tool Name AccuracyLlama3-70B
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| Llama3-70BModel Series=Llama3 Series, Method=HiTEC-KTO2025.05 | 91.17 | 60.83 | |
| Qwen2.5-7BModel Series=Qwen2.5 Series, Method=HiTEC-KTO2025.05 | 87.63 | 59.19 | |
| HiTEC-KTOModel=Qwen2.5-7B2025.05 | 87.63 | 59.19 | |
| HiTEC-ICLModel=GPT-4-Turbo2025.05 | 87.35 | 61.95 | |
| HiTEC-ICLBackbone=GPT-4-Turbo2025.05 | 87.35 | 61.95 | |
| HiTEC-ICLModel=GPT-3.5-Turbo-01252025.05 | 87.27 | 58.79 | |
| VanillaBackbone=GPT-4-Turbo2025.05 | 87.12 | 60.75 | |
| Function CallingBackbone=GPT-4-Turbo2025.05 | 87.11 | 62.16 | |
| Llama3-8BModel Series=Llama3 Series, Method=HiTEC-KTO2025.05 | 87.02 | 52.4 | |
| HiTEC-KTOModel=Llama3-8B2025.05 | 87.02 | 52.4 | |
| Llama3-70BModel Series=Llama3.1 Series, Method=HiTEC-KTO2025.05 | 86.67 | 57.32 | |
| Llama3.1-70BModel Series=Llama3.1 Series, Method=Baseline2025.05 | 86.03 | 55.83 | |
| HiTEC-ICLModel=Llama3-8B2025.05 | 85.92 | 55.49 | |
| HiTEC-ICLBackbone=Llama3-8B2025.05 | 85.92 | 55.49 | |
| HiTEC-ICLBackbone=Llama3.1-8B2025.05 | 85.92 | 55.94 | |
| Qwen2.5-3BModel Series=Qwen2.5 Series, Method=HiTEC-KTO2025.05 | 85.42 | 56.08 | |
| Llama3.1-8BModel Series=Llama3.1 Series, Method=HiTEC-KTO2025.05 | 84.18 | 56.45 | |
| Llama3.1-8BModel Series=Llama3.1 Series, Method=Baseline2025.05 | 84.06 | 50.79 | |
| VanillaBackbone=Llama3.1-8B2025.05 | 84.06 | 50.79 | |
| Qwen2.5-1.5BModel Series=Qwen2.5 Series, Method=HiTEC-KTO2025.05 | 82.27 | 52.26 | |
| Hammer2-7BModel Series=Qwen2.5 Series, Method=Baseline2025.05 | 81.74 | 57.07 | |
| CoTBackbone=GPT-4-Turbo2025.05 | 80.95 | 56.46 | |
| Hammer2-1.5BModel Series=Qwen2.5 Series, Method=Baseline2025.05 | 80.74 | 51.88 | |
| Hammer2-3BModel Series=Qwen2.5 Series, Method=Baseline2025.05 | 78.23 | 53.28 | |
| Qwen2.5-0.5BModel Series=Qwen2.5 Series, Method=HiTEC-KTO2025.05 | 73.54 | 38.76 | |
| Hammer2-0.5BModel Series=Qwen2.5 Series, Method=Baseline2025.05 | 64.46 | 41.83 | |
| CoTBackbone=Llama3.1-8B2025.05 | 63.31 | 42.57 | |
| CoTBackbone=Llama3-8B2025.05 | 60.58 | 42.07 | |
| Llama3-70BModel Series=Llama3 Series, Method=Baseline2025.05 | 49.64 | 47.73 | |
| Llama3-8BModel Series=Llama3 Series, Method=Baseline2025.05 | 40.46 | 24.86 | |
| VanillaBackbone=Llama3-8B2025.05 | 40.46 | 24.86 |