LLM Cube Stacking Project

Undergraduate Research Project: To assess the capabilities of LLM and VLM in cube stacking for common-sense planning challenges. With the goal of generating optimized sequences of events to address cube-stacking challenges using openai
Explore the docs »

View Demo · Report Bug · Request Feature

As a UCF undergraduate researcher, I am utilizing the OpenAI, also known as ChatGPT-4 API, in conjunction with Python scripts to assess and compare its capabilities in common sense reasoning with ChatGPT-3.5. My undergraduate research project focuses on assessing the capabilities of LLM (Large Language Model) and VLM (Very Large Language Model) in cube stacking for common-sense planning challenges. The primary objective of this project is to generate optimized sequences of events to address cube-stacking challenges using OpenAI, particularly in the context of the Cube Stacking problem, which is one of the most renowned planning domains in the realm of artificial intelligence. In this challenge, the goal is to construct one or more vertical stacks of blocks to transform the initial state into the goal state. Notably, only one block may be moved at a time, and it can be placed either on the table or on top of another block, with the condition that a block cannot be moved if another block is resting on it.

• npm

  npm install npm@latest -g

1. Clone the repo:

   git clone https://github.com/christina8711/llm_cubestacking.git

   • This step clones the repository from GitHub to your local machine.

2. Navigate to the project directory:

   cd llm_cubestacking

   • Change your working directory to the newly cloned repository.

3. Install Node.js and npm (if not already installed):

   • Ensure that Node.js and npm are installed on your system. You can download them from the official Node.js website.

4. Install project-specific dependencies:

   npm install

   • Use this command to install the project-specific Node.js packages and dependencies listed in the package.json file.

(back to top)

• Create a controlled environment, simulating a real-world 'table' scenario, for ChatGPT-4 to operate within.
• Scale up the cube stacking problem to further challenge ChatGPT-4's common sense reasoning capabilities.
  • This includes increasing the number of blocks in the problem and introducing random changes to the controlled environment.
• Provide feedback to ChatGPT-4 in case of an error.
  • If ChatGPT's response is incorrect, communicate that a mistake was made.
  • Try again if needed.

See the open issues for a full list of proposed features (and known issues).

(back to top)

Contributions are what makes the open source community such an amazing place to learn, inspire, and create. Any contributions you make are greatly appreciated.

If you have a suggestion that would make this better, please fork the repo and create a pull request. You can also simply open an issue with the tag "enhancement". Don't forget to give the project a star! Thanks again!

1. Fork the Project
2. Create your Feature Branch (git checkout -b feature/AmazingFeature)
3. Commit your Changes (git commit -m 'Add some AmazingFeature')
4. Push to the Branch (git push origin feature/AmazingFeature)
5. Open a Pull Request

(back to top)

Christina Nguyen - - ch487674@ucf.edu

Project Link: https://github.com/christina8711/llm_cubstacking

(back to top)