Example scripts and configurations for text-to-image experiments are collected in scripts/exp-text2img and configs/exp-text2img respectively.
Here we present a brief tutorial on how to run experiments with these scripts and configurations (Please check comments in these scripts to see how to set encoder, downstream task, etc.).
Step 0: Enter the folder:
cd ./scripts/exp-text2img
Step 1: Synthesize PEI attack texts for a candidate encoder (e.g., clip-vit-l14) with the configuration configs/exp-text2img/atk.yaml (PEI objective texts are presented in the script):
bash atk.sh ../../
The used PEI objective texts are generated via scripts/gen_obj_text.py.
Step 2: Calculate the PEI score of a candidate encoder (e.g., clip-vit-b16 with the configuration configs/exp-text2img/eval-b16.yaml) on a targeted downstream text-to-image service (e.g., sd-v1-2):
bash eval.sh ../../
Obtained PEI scores can then be used to calculate PEI z-scores to conduct the PEI attack (see the paper for details).