Safe Autoregressive Image Generation with Iterative Self-Improving Codebooks
安全自回归图像生成与迭代自改进码本
机构 * School of Information Management, Wuhan University(武汉大学信息管理学院) ; Torr Vision Group, University of Oxford(牛津大学Torr视觉组)
专题命中 多模态生成 :multimodal(abstract);image-text(abstract);分类 cs.CV、cs.AI
AI总结 针对自回归统一多模态模型生成图像的安全性问题,提出迭代自改进码本方法,利用模型自身判断不安全图像并修正码本映射,消除有害输出并保证生成质量。
Comments 10 pages including references, 8 figures, accepted for publication at the 43rd International Conference on Machine Learning (ICML 2026)