OneReward:基于多任务人类偏好学习的统一掩码引导图像生成仅供参考,未经实验验证。论文标题:OneReward: Unified Mask-Guided Image Generation via Multi-Task Human Preference Learning 作者:Yuan Gong, Xionghui Wang, Jie Wu, Shiyin Wang, Yitong Wang, Xinglong Wu 机构:字节跳动 论文地址: https://arxiv.org/pdf/2508.21066 Github地址:https://github.com