Baselines

Proposed Model

Pseudo_code

Experimental Results

How to run the frameworks

Individual Frameworks:

To run individual frameworks, follow these steps:

1. Navigate to the directory:

cd code/GPT3.5/individual_frameworks/LLM_KG_Human

2. Run the script using the following format:

python3 stitching_everything_exp.py <input-prompt> <no-of-feedback> <state-file>

3. Example command:

python3 stitching_everything_exp.py "Give me cot and action sequence to prepare an omlette" 3 "intial_onto.ttl"

4. Adjust the input-prompt, no-of-feedback, and state-file parameters as needed for the specific case.

Streamlined Frameworks:

To run the streamlined frameworks, follow these steps:

1. Navigate to the directory:

cd code/GPT3.5/cooking_domain

2. Run the script using the following format:

python3 stitching_everything.py <no-of-feedback> <state-file>

3. Note: All tasks are listed in the kitchen.json file located in the code/GPT3.5/cooking_domain directory.

4. Example command:

python3 stitching_everything.py 3 "initial_onto.ttl"

Installing Dependencies:

You can install the dependencies using the following command:

pip install -r requirements.txt