扩散模型中降低幻觉的分数控制

摘要

扩散模型已成为现代生成式AI的基石，推动着视觉、语言、音频及其他模态的进步。尽管取得了成功，这些模型仍存在幻觉问题——即生成超出真实数据分布支持范围的不可信样本，这降低了可靠性与可信度。在本工作中，我们首先通过实验验证了先前提出的假设：图像生成扩散模型中分数平滑性会导致幻觉，并提供了基于密度的视角。我们进一步将这一概念形式化，通过将幻觉概率质量与所学分数函数的Lipschitz常数建立关联。受此启发，我们提出一种方差引导的分数调制（VSM）策略，通过控制分数雅可比矩阵来降低分数平滑性，从而更准确地逼近真实分数，减少幻觉现象。在合成数据集与真实世界数据集上的实验结果表明，我们的方法可在保持高保真度与多样性的同时，将幻觉降低约25%，为构建更可靠的基于扩散的图像生成模型提供了原则性步骤。此外，我们还提出了两个具有极端语义变化的基准数据集，用于系统性评估幻觉。代码与数据集已公开于 https://github.com/bhosalems/VSM。

English

Diffusion models have emerged as the backbone of modern generative AI, powering advances in vision, language, audio and other modalities. Despite their success, they suffer from hallucinations, implausible samples that lie outside the support of true data distribution, which degrade reliability and trust. In this work, we first empirically confirm previously proposed hypothesis that score smoothness causes hallucinations in Image Generation diffusion models and provide a density-based perspective. We further formalize this notion by linking the hallucinations probability mass to lipschitz constant of the learned score function. Motivated by this, we introduce a Variance-Guided Score Modulation (VSM) strategy that controls the score Jacobian, in turn reducing score smoothness and better approximating the ground truth score that decreases hallucinations. Empirical results on synthetic and real-world datasets demonstrate that our approach reduces hallucinations (up to ~25%) while maintaining high fidelity and diversity, providing a principled step toward more reliable diffusion-based image generation. We also propose two benchmark datasets with extreme semantic variation for systematic hallucination evaluation. Code and Datasets are publicly available at https://github.com/bhosalems/VSM.