ToolBench is an instruction-tuning dataset for tool use, which is created automatically using ChatGPT. Specifically, the authors collect 16,464 real-world RESTful APIs spanning 49 categories from RapidAPI Hub, then prompt ChatgPT to generate diverse human instructions involving these APIs, covering both single-tool and multi-tool scenarios.
Variants: ToolBench
This dataset is used in 1 benchmark:
| Task | Model | Paper | Date |
|---|---|---|---|
| Trajectory Planning | GPT4-TOPGUN | SwissNYF: Tool Grounded LLM Agents … | 2024-02-15 |
| Trajectory Planning | Attention Bucket | Fortify the Shortest Stave in … | 2023-12-07 |
| Trajectory Planning | GPT4- DFSDT | ToolLLM: Facilitating Large Language Models … | 2023-07-31 |
Recent papers with results on this dataset: