Agentic-CLIP-Benchmark

Agentic-CLIP-Benchmark: Evaluación Zero-Shot en CIFAR-10 -green.svg Este proyecto implementa un pipeline robusto y automatizado para evaluar el modelo CLIP ViT-B/32 de OpenAI en el conjunto de prueba completo de CIFAR-10 10,000 imágenes. Desarrollado mediante un flujo de trabajo AI-Native Trae ID...

Enhancing One-run Privacy Auditing with Quantile Regression-Based Membership Inference

Differential privacy DP auditing aims to provide empirical lower bounds on the privacy guarantees of DP mechanisms like DP-SGD. While some existing techniques require many training runs that are prohibitively costly, recent work introduces one-run auditing approaches that effectively audit DP-SGD...