MoESD:用于缓解性别偏见的混合专家 Stable Diffusion
MoESD: Mixture of Experts Stable Diffusion to Mitigate Gender Bias
AI总结:
针对文本到图像模型的性别偏见,本文提出带偏见适配器与偏见识别门的 MoESD,在潜在空间识别文本编码偏见并借助特殊提示标记缓解偏见,且保持图像质量。
摘要:
文本到图像模型已知会传播社会偏见。例如,当被提示生成某些职业人物的图像时,这些模型往往会系统性地生成特定性别或族裔。在本文中,我们表明这种偏见已经存在于模型的文本编码器中,并通过在潜在空间中识别文本编码的偏见,随后创建偏见识别门机制,引入了一种混合专家方法。更具体地说,我们提出带有 BiAs(偏见适配器)的 MoESD(混合专家 Stable Diffusion),以缓解文本到图像模型中的性别偏见。我们还证明,在缓解过程中,向提示引入任意特殊标记至关重要。通过聚焦性别偏见的实验,我们表明我们的方法成功缓解了性别偏见,同时保持了图像质量。
英文摘要:
Text-to-image models are known to propagate social biases. For example, when prompted to generate images of people in certain professions, these models tend to systematically generate specific genders or ethnicities. In this paper, we show that this bias is already present in the text encoder of the model and introduce a Mixture-of-Experts approach by identifying text-encoded bias in the latent space and then creating a Bias-Identification Gate mechanism. More specifically, we propose MoESD (Mixture of Experts Stable Diffusion) with BiAs (Bias Adapters) to mitigate gender bias in text-to-image models. We also demonstrate that introducing an arbitrary special token to the prompt is essential during the mitigation process. With experiments focusing on gender bias, we show that our approach successfully mitigates gender bias while maintaining image quality.
机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。