Tool Calling on Average across 5 benchmarks
88.47F1 (Name)HiTEC-ICL
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| HiTEC-ICLBackbone=GPT-4-Turbo2025.05 | 88.47 | 79.12 | |
| VanillaBackbone=GPT-4-Turbo2025.05 | 88.39 | 77.86 | |
| HiTEC-ICLBackbone=Llama3.1-8B2025.05 | 85.59 | 66.42 | |
| CoTBackbone=GPT-4-Turbo2025.05 | 84.84 | 75.62 | |
| VanillaBackbone=Llama3.1-8B2025.05 | 83.65 | 64.67 | |
| HiTEC-ICLBackbone=Llama3-8B2025.05 | 83.12 | 65.52 | |
| CoTBackbone=Llama3.1-8B2025.05 | 73.87 | 60.08 | |
| CoTBackbone=Llama3-8B2025.05 | 59.49 | 52.03 | |
| VanillaBackbone=Llama3-8B2025.05 | 57.31 | 48.29 |