Hugging Face

Hugging Face is a platform that provides pre-trained machine learning models in standard formats to facilitate accessibility and reuse, and it is used to evaluate the detection of malicious models in the context of machine learning supply chain security.

Papers using Hugging Face (3)

Your Compiler is Backdooring Your Model: Understanding and Exploiting Compilation Inconsistency Vulnerabilities in Deep Learning Compilers2025 · 1 citesMalicious ML Model Detection by Learning Dynamic Behaviors2026SafePickle: Robust and Generic ML Detection of Malicious Pickle-based ML Models2026