CoolFace
Datasetpublic

closerG/ppu-bench

PPU-Bench: Real-World Multimodal Benchmark for Personalized Partial Unlearning PPU-Bench is a real-world multimodal benchmark designed to evaluate personalized partial unlearning in vision-language models. It supports multiple unlearning settings and provides training/evaluation data for different VLM backbones.

sourceHugging Facemitupdated 5mo agoView on Hugging Face
0likes568downloads

No commit history came back for main. The revision may not exist, or the source declined the request.