|
--- |
|
dataset_info: |
|
features: |
|
- name: url |
|
dtype: string |
|
- name: caption |
|
dtype: string |
|
splits: |
|
- name: members |
|
num_bytes: 2192248 |
|
num_examples: 13396 |
|
- name: nonmembers |
|
num_bytes: 4395869 |
|
num_examples: 26874 |
|
download_size: 5044720 |
|
dataset_size: 6588117 |
|
license: mit |
|
task_categories: |
|
- text-to-image |
|
language: |
|
- en |
|
tags: |
|
- membership |
|
- latent diffusion |
|
pretty_name: LAION-mi |
|
size_categories: |
|
- 10K<n<100K |
|
--- |
|
# Dataset Card for "LAION-mi" |
|
|
|
This is a dataset from Towards More Realistic Membership Inference Attacks on Large Diffusion Models, [link](https://arxiv.org/abs/2306.12983). |
|
|
|
We provide a robust evaluation setup for membership inference attacks on a state-of-the-art Stable Diffusion model and publish the corresponding dataset along with the proposed setup. |
|
|
|
# Citing |
|
|
|
If you find our work useful in your research, please cite as |
|
|
|
``` |
|
@misc{dubiński2023realistic, |
|
title={Towards More Realistic Membership Inference Attacks on Large Diffusion Models}, |
|
author={Jan Dubiński and Antoni Kowalczuk and Stanisław Pawlak and Przemysław Rokita and Tomasz Trzciński and Paweł Morawiecki}, |
|
year={2023}, |
|
eprint={2306.12983}, |
|
archivePrefix={arXiv}, |
|
primaryClass={cs.LG} |
|
} |
|
``` |