Skip to content

Characterising the Creative Process in Humans and Large Language Models

Notifications You must be signed in to change notification settings

surabhisnath/Creative_Process

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

15 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Characterising the Creative Process in Humans and Large Language Models

python jupyter

This repository contains the data and scripts for the paper Characterising the Creative Process in Humans and Large Language Models accepted at ICCC'24. Authors: Surabhi S Nath, Peter Dayan, Claire Stevenson

Abstract

Large language models appear quite creative, often performing on par with the average human on creative tasks. However, research on LLM creativity has focused solely on products, with little attention on the creative process. Process analyses of human creativity often require hand-coded categories or exploit response times, which do not apply to LLMs. We provide an automated method to characterise how humans and LLMs explore semantic spaces on the Alternate Uses Task, and contrast with behaviour in a Verbal Fluency Task. We use sentence embeddings to identify response categories and compute semantic similarities, which we use to generate jump profiles. Our results corroborate earlier work in humans reporting both persistent (deep search in few semantic spaces) and flexible (broad search across multiple semantic spaces) pathways to creativity, where both pathways lead to similar creativity scores. LLMs were found to be biased towards either persistent or flexible paths, that varied across tasks. Though LLMs as a population match human profiles, their relationship with creativity is different, where the more flexible models score higher on creativity.

Graphical Abstract

Repository Description

Human data files in data
LLM data files in data_LLM
Scripts (and requirements) to call LLMs in scripts_LLM
Response embeddings in embeddings
Response translations in translations
Paper figures in figures
Script to make csvs in scripts/make_csvs.ipynb. Final csv provided in csvs/ as data_humans.csv and data_LLMs.csv
All analyses in scripts/Analysis.ipynb. This is the only file which needs to be run to obtain all plots and results.

Setup

We recommend setting up a python virtual environment and installing all the requirements. Please follow these steps:

git clone https://github.com/surabhisnath/Creative_Process.git
cd Creative_Process

python3 -m venv .env

# On macOS/Linux
source .env/bin/activate
# On Windows
.env\Scripts\activate

pip install -r requirements.txt

Running the code

To reproduce the results from the paper, run

jupyter-lab

and open the file scripts/Analysis.ipynb and run the file (either in one go or cell by cell).
If you wish to re-create the csv file, run scripts/make_csvs.ipynb.

Citation

If you found this work useful, please consider citing us:

@article{nath2024characterising,
  title={Characterising the Creative Process in Humans and Large Language Models},
  author={Nath, Surabhi S and Dayan, Peter and Stevenson, Claire},
  journal={arXiv preprint arXiv:2405.00899},
  year={2024}
}

Contact

Please feel free to email us at nath.surabhi@tuebingen.mpg.de.