https://github.com/quchangle1/DRAFT

The implementation for the paper: From Exploration to Mastery: Enabling LLMs to Master Tools via Self-Driven Interactions.
https://github.com/quchangle1/DRAFT

Last synced: 3 days ago
JSON representation

The implementation for the paper: From Exploration to Mastery: Enabling LLMs to Master Tools via Self-Driven Interactions.

• Host: GitHub
• URL: https://github.com/quchangle1/DRAFT
• Owner: quchangle1
• Created: 2024-10-02T06:17:18.000Z (3 months ago)
• Default Branch: main
• Last Pushed: 2024-12-28T14:24:03.000Z (11 days ago)
• Last Synced: 2024-12-28T15:22:30.116Z (11 days ago)
• Language: Python
• Homepage:
• Size: 404 KB
• Stars: 7
• Watchers: 1
• Forks: 0
• Open Issues: 1
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

• awesome_ai_agents - Draft - The implementation for the paper - From Exploration to Mastery - Enabling LLMs to Master Tools via Self-Driven Interactions. (Building / Tools)
• awesome_ai_agents - Draft - The implementation for the paper - From Exploration to Mastery - Enabling LLMs to Master Tools via Self-Driven Interactions. (Building / Tools)

README

        
# DRAFT

The implementation for the paper: From Exploration to Mastery: Enabling LLMs to Master Tools via Self-Driven Interactions.

## How to run the code

1. Download ToolBench dataset from the [Google Drive](https://drive.google.com/drive/folders/1yBUQ732mPu-KclJnuQELEhtKakdXFc3J).

2. Run DRAFT to get revised tool documentation:

	> python DRAFT.py

3. Run ToolBench_DFSDT to examine the effectiveness of DRAFT:

	> python ToolBench_DFSDT -model_name gpt-4o-2024-08-06 -data_type G3 -method DRAFT

## Environment

Our experimental environment is shown below:

```

openai version: 0.28.0

numpy version: 1.26.4

pandas version: 2.2.2

torch version: 2.3.1

```