DreamBench++：一个针对个性化图像生成的人类对齐基准。

DreamBench++: A Human-Aligned Benchmark for Personalized Image Generation

June 24, 2024

作者: Yuang Peng, Yuxin Cui, Haomiao Tang, Zekun Qi, Runpei Dong, Jing Bai, Chunrui Han, Zheng Ge, Xiangyu Zhang, Shu-Tao Xia

cs.AI

摘要

个性化图像生成在辅助人类日常工作和生活方面具有巨大潜力，因为它在创造性生成个性化内容方面具有令人印象深刻的功能。然而，当前的评估要么是自动化的但与人类不一致，要么需要耗时且昂贵的人类评估。在这项工作中，我们提出了DreamBench++，这是一个由先进的多模态GPT模型自动化的与人类一致的基准。具体来说，我们系统地设计提示，让GPT既与人类一致又自我一致，并赋予其任务强化能力。此外，我们构建了一个包含多样化图像和提示的全面数据集。通过对7种现代生成模型进行基准测试，我们证明DreamBench++导致了显著更多与人类一致的评估结果，有助于推动社区获得创新性发现。

English

Personalized image generation holds great promise in assisting humans in everyday work and life due to its impressive function in creatively generating personalized content. However, current evaluations either are automated but misalign with humans or require human evaluations that are time-consuming and expensive. In this work, we present DreamBench++, a human-aligned benchmark automated by advanced multimodal GPT models. Specifically, we systematically design the prompts to let GPT be both human-aligned and self-aligned, empowered with task reinforcement. Further, we construct a comprehensive dataset comprising diverse images and prompts. By benchmarking 7 modern generative models, we demonstrate that DreamBench++ results in significantly more human-aligned evaluation, helping boost the community with innovative findings.