Abstract A label‐free electrochemical method based on scanning electrochemical microscopy (SECM) has been developed to image latent fingerprints with high resolution on five kinds of metal surfaces (platinum, gold, silver, copper and stainless steel), as it could measure the minor conductivity differences of the substrate surface and avoid the interference of the background‐color. The images of sebaceous fingerprints on clean metals were revealed by SECM with ferrocene methanol acting as a redox mediator to detect the topology of the fingerprint deposits in constant‐height feedback mode. Inhibition of electrochemical processes on areas of the surface masked by the insulating fingerprint residues generated a negative image of the fingerprint.
Freckle defects were found in electro-slag remelting (ESR) ingots, which severely limit the development of the larger-diameter ingots. The macrostructure and the composition of the freckles in a low-alloy structure steel ESR ingot were investigated to clarify the freckle formation mechanism. The results shows that the freckles in the ingot were caused by the upward solute-rich liquid flow in the mushy zone, and the composition of the freckles corresponds to that of the liquid, with a liquid fraction of 0.36 to 0.43. The relative Rayleigh number (Ra), a freckle criterion considering the effect of a tilted solidification front, was adopted to evaluate the freckle formation tendency in the industrial-scale ESR ingots based on the results of the experiments, kinetic simulation and thermophysical property calculation. The calculated results of Ra are in good agreement with the actual distribution of the freckles in the ingot.
Generative models have demonstrated revolutionary success in various visual creation tasks, but in the meantime, they have been exposed to the threat of leaking private information of their training data. Several membership inference attacks (MIAs) have been proposed to exhibit the privacy vulnerability of generative models by classifying a query image as a training dataset member or nonmember. However, these attacks suffer from major limitations, such as requiring shadow models and white-box access, and either ignoring or only focusing on the unique property of diffusion models, which block their generalization to multiple generative models. In contrast, we propose the first generalized membership inference attack against a variety of generative models such as generative adversarial networks, [variational] autoencoders, implicit functions, and the emerging diffusion models. We leverage only generated distributions from target generators and auxiliary nonmember datasets, therefore regarding target generators as black boxes and agnostic to their architectures or application scenarios. Experiments validate that all the generative models are vulnerable to our attack. For instance, our work achieves attack AUC > 0.99 against DDPM, DDIM, and FastDPM trained on CIFAR-10 and CelebA. And the attack against VQGAN, LDM (for the text-conditional generation), and LIIF achieves AUC > 0.90. As a result, we appeal to our community to be aware of such privacy leakage risks when designing and publishing generative models. 1
'/%3e%3cuse xlink:href='%23a' transform='rotate(23.036 .093 25.536)'/%3e%3cuse xlink:href='%23a' transform='rotate(45.87 1.273 16.18)'/%3e%3cuse xlink:href='%23a' transform='rotate(69.945 .996 12.078)'/%3e%3cuse xlink:href='%23a' transform='rotate(20.66 -19.689 31.932)'/%3e%3c/svg%3e)
