Advisories for Pypi/Pytorch package

A vulnerability in PyTorch's weights_only unpickler allows an attacker to craft a malicious checkpoint file (.pth) that, when loaded with torch.load(…, weights_only=True), can corrupt memory and potentially lead to arbitrary code execution.

I found a Remote Command Execution (RCE) vulnerability in PyTorch. When loading model using torch.load with weights_only=True, it can still achieve RCE.