通过稀疏监督掩码提升扩散模型生成一致性
Sparsely Supervised Diffusion
- 训练时随机掩蔽高达98%像素,仅需少量代码
- 在多个数据集上实现媲美主流方法的FID分数
- 特别适合小数据集,减少记忆过拟合
扩散模型在各类生成任务中表现卓越,但常因去噪机制的局部性导致空间不一致,生成结果局部合理却全局矛盾。为此,我们提出稀疏监督扩散学习,一种简单有效的掩码策略,仅需几行代码即可实现。实验表明,训练时掩蔽高达98%像素仍安全有效。该方法在多项实验中取得竞争力的FID分数,尤其在小数据集上避免了训练不稳定问题。同时,掩码策略降低了模型对训练样本的记忆,促进生成时利用关键上下文信息。
原文摘要 · Abstract (English)
Diffusion models have shown remarkable success across a wide range of generative tasks. However, they often suffer from spatially inconsistent generation, arguably due to the inherent locality of their denoising mechanisms. This can yield samples that are locally plausible but globally inconsistent. To mitigate this issue, we propose sparsely supervised learning for diffusion models, a simple yet effective masking strategy that can be implemented with only a few lines of code. Interestingly, the experiments show that it is safe to mask up to 98\% of pixels during diffusion model training. Our method delivers competitive FID scores across experiments and, most importantly, avoids training instability on small datasets. Moreover, the masking strategy reduces memorization and promotes the use of essential contextual information during generation.
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。