README.md

## File Descriptions

### `main.py`

This is the main script that sets up the entire process, including loading datasets, model initialisation, and evaluation.

To execute the main script, use the following command:

```bash
cd code_question
```

```bash
python main.py --dataset  mmlu-redux-college_mathematics  --method comat --model gpt
```

### `mmlu_redux.py`

This file contains functions for processing the MMLU-Redux dataset.

- **`process_mmlu_redux_questions`**: Processes questions from the MMLU-Redux dataset and evaluates them.

This directory also contains prompt files associated with the College Mathematics dataset:

- **`MMLU-Redux-college_mathematics_prompts/comat.txt`**: Prompt file for the CoMAT method.
- **`CoMAT_Instruction.py`**: Prompt file for the CoMAT method, used for evaluation in this assignment.


### `shapley_value_evaluation.py`

This script evaluates the Shapley value, a concept from cooperative game theory. The Shapley value is used to fairly distribute total gains among participants based on their individual contributions.

### `utils.py`

This file contains functions for using gpt to make predictions with different models and evaluate their outputs.