Image Corruption-Inspired Membership Inference Attacks against Large Vision-Language Models
By: Zongyu Wu , Minhua Lin , Zhiwei Zhang and more
Potential Business Impact:
Finds if your pictures trained AI.
Large vision-language models (LVLMs) have demonstrated outstanding performance in many downstream tasks. However, LVLMs are trained on large-scale datasets, which can pose privacy risks if training images contain sensitive information. Therefore, it is important to detect whether an image is used to train the LVLM. Recent studies have investigated membership inference attacks (MIAs) against LVLMs, including detecting image-text pairs and single-modality content. In this work, we focus on detecting whether a target image is used to train the target LVLM. We design simple yet effective Image Corruption-Inspired Membership Inference Attacks (ICIMIA) against LLVLMs, which are inspired by LVLM's different sensitivity to image corruption for member and non-member images. We first perform an MIA method under the white-box setting, where we can obtain the embeddings of the image through the vision part of the target LVLM. The attacks are based on the embedding similarity between the image and its corrupted version. We further explore a more practical scenario where we have no knowledge about target LVLMs and we can only query the target LVLMs with an image and a question. We then conduct the attack by utilizing the output text embeddings' similarity. Experiments on existing datasets validate the effectiveness of our proposed attack methods under those two different settings.
Similar Papers
OpenLVLM-MIA: A Controlled Benchmark Revealing the Limits of Membership Inference Attacks on Large Vision-Language Models
CV and Pattern Recognition
Finds if AI remembers private training pictures.
Black-Box Membership Inference Attack for LVLMs via Prior Knowledge-Calibrated Memory Probing
Cryptography and Security
Uncovers if AI saw your private pictures.
Membership Inference Attacks on Large-Scale Models: A Survey
Machine Learning (CS)
Finds if your private info trained AI.