sayzardSaar Huberman (@HubermanSaar)
‘Image Generation from Contextually Contradictory Prompts’ 연구가 CVPR 2026에 채택됐다. 확산 모델이 겉보기엔 단순한 프롬프트에서도 지시를 무시하는 이유를 분석하고, 이를 훈련 없이 해결하는 방법을 제시한다.
Saar Huberman (@HubermanSaar) on X
Excited to share that our work, “Image Generation from Contextually Contradictory Prompts,” was accepted to CVPR 2026 🎉 Diffusion models fail on some seemingly simple prompts. Why do they ignore what you asked for? We show why, and how to fix it with a simple, training-free