This is the official implementation of the CVPR 2026 paper: "Scaling Test-Time Robustness of Vision-Language Models via Self-Critical Inference Framework" - View it on GitHub