Ludwig Framework Insecure Deserialization Vulnerability in Model Serving Component

A vulnerability allowing insecure deserialization has been identified in the Ludwig framework, affecting versions through 0.10.4. The issue arises in the model serving component, where the framework loads model weight files using torch.load() without the security-restrictive weights_only=True parameter. This oversight enables the deserialization of arbitrary Python objects via the pickle module. An attacker could exploit this vulnerability by providing a maliciously crafted PyTorch model file, potentially leading to arbitrary code execution on the system hosting the Ludwig model server.

Impact

Exploitation of this vulnerability could result in arbitrary code execution on the server where Ludwig is running.