Tool Learning on StableToolBench I1-Tool
73.9SoPRGPT-3.5 (DFSDT)
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| GPT-3.5 (DFSDT)Backbone=GPT-3.5, Reasoning Strategy=DFSDT2025.01 | 73.9 | 65.8 | |
| GPT-4 (DFSDT)Backbone=GPT-4, Reasoning Strategy=DFSDT2025.01 | 69.6 | 66.5 | |
| GPT-4 (Parallel)Backbone=GPT-4, Reasoning Strategy=Parallel2025.01 | 67.4 | 61.4 | |
| GPT-3.5 (Parallel)Backbone=GPT-3.5, Reasoning Strategy=Parallel2025.01 | 65 | 55.7 | |
| DTA-LlamaBackbone=Llama-2-7B, Reasoning Strategy=Parallel, Training Protocol=Fine-tuned on DTA-Tool2025.01 | 64.2 | 53.2 | |
| Qwen2.5 (Parallel)Backbone=Qwen2.5-7B-Instruct, Reasoning Strategy=Parallel2025.01 | 58.8 | 51 | |
| ToolLLAMA (DFSDT)Backbone=Llama-2-7B, Reasoning Strategy=DFSDT2025.01 | 55.5 | 46.8 | |
| GPT-3.5 (ReAct)Backbone=GPT-3.5, Reasoning Strategy=ReAct2025.01 | 53 | — | |
| GPT-4 (ReAct)Backbone=GPT-4, Reasoning Strategy=ReAct2025.01 | 44.1 | 60.1 | |
| ToolLLaMA† (DFSDT)Backbone=Llama-2-7B, Reasoning Strategy=DFSDT, Training Protocol=DTA-Tool*2025.01 | 39.9 | 37.3 | |
| ToolLLAMA (ReAct)Backbone=Llama-2-7B, Reasoning Strategy=ReAct2025.01 | 35.4 | 36.1 | |
| LLMCompilerBackbone=Llama-2-7B, Reasoning Strategy=Parallel2025.01 | 35.1 | 36 | |
| ToolLLaMA† (ReAct)Backbone=Llama-2-7B, Reasoning Strategy=ReAct, Training Protocol=DTA-Tool*2025.01 | 25 | 27.2 |