ImagenWorld: Stress-Testing Image Generation Models with Explainable Human Evaluation on Open-ended Real-World Tasks Paper • 2603.27862 • Published 3 days ago • 21
Usefulness Judge Collection Finetuned judges to evaluate how useful a response is to a prompt • 5 items • Updated Feb 4
Expected Harm: Rethinking Safety Evaluation of (Mis)Aligned LLMs Paper • 2602.01600 • Published Feb 2 • 21