流匹配模型在数据与架构变动下仍保持生成质量与多样性。
The Amazing Stability of Flow Matching
- 通过剪枝50%数据集，模型生成质量与多样性不变
- 剪枝后隐空间表示变化极小，同种子输出视觉相似
- 适用于对模型鲁棒性要求高的生成任务
深度生成模型生成高质量、多样化样本的成功常归因于特定架构和大规模训练数据。本文研究这些因素对流匹配模型生成质量与多样性的影 响。在CelebA-HQ数据集上，令人惊讶的是，即使剪枝50%数据，流匹配模型仍保持稳定：生成样本的质量与多样性得以保留。此外，剪枝仅轻微影响隐空间表示，即在完整与剪枝数据集上训练的模型对同一随机种子生成的输出视觉相似。我们还发现，改变架构或训练配置时，隐空间表示同样保持稳定。结果量化了实际中这种稳定性程度，有助于解释流匹配模型在各种扰动下的可靠性。
原文摘要 · Abstract (English)
The success of deep generative models in generating high-quality and diverse samples is often attributed to particular architectures and large training datasets. In this paper, we investigate the impact of these factors on the quality and diversity of samples generated by \emph{flow-matching} models. Surprisingly, in our experiments on CelebA-HQ dataset, flow matching remains stable even when pruning 50\% of the dataset. That is, the quality and diversity of generated samples are preserved. Moreover, pruning impacts the latent representation only slightly, that is, samples generated by models trained on the full and pruned dataset map to visually similar outputs for a given seed. We observe similar stability when changing the architecture or training configuration, such that the latent representation is maintained under these changes as well. Our results quantify just how strong this stability can be in practice, and help explain the reliability of flow-matching models under various perturbations.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。