用可学习的噪声-数据关联提升一致性模型训练稳定性
VCT: Training Consistency Models with Variational Noise Coupling
- 引入变分噪声耦合机制，让模型自适应学习噪声与数据配对
- 在CIFAR-10上达到非蒸馏一致性模型最佳FID，ImageNet仅需2步采样即达顶尖性能
- 适用于多种前向核，尤其适配流匹配方法，适合追求高效生成的研究者
一致性训练（CT）近年来成为图像生成的有力替代方案，但非蒸馏CT常因高方差和不稳定性而受限。本文提出变分一致性训练（VCT），一种兼容多种前向核（包括流匹配）的灵活有效框架。其核心创新在于受变分自编码器启发的可学习噪声-数据耦合机制，通过数据相关编码器建模噪声发射过程。该机制使VCT能自适应学习噪声与数据的配对关系，显著降低经典CT中固定、无序配对带来的训练方差。在多个图像数据集上的实验表明，本方法性能显著优于基线，在CIFAR-10上实现非蒸馏CT的最佳FID，且在ImageNet 64×64上仅用两步采样即达到当前最优表现。代码已开源。
原文摘要 · Abstract (English)
Consistency Training (CT) has recently emerged as a strong alternative to diffusion models for image generation. However, non-distillation CT often suffers from high variance and instability, motivating ongoing research into its training dynamics. We propose Variational Consistency Training (VCT), a flexible and effective framework compatible with various forward kernels, including those in flow matching. Its key innovation is a learned noise-data coupling scheme inspired by Variational Autoencoders, where a data-dependent encoder models noise emission. This enables VCT to adaptively learn noise-todata pairings, reducing training variance relative to the fixed, unsorted pairings in classical CT. Experiments on multiple image datasets demonstrate significant improvements: our method surpasses baselines, achieves state-of-the-art FID among non-distillation CT approaches on CIFAR-10, and matches SoTA performance on ImageNet 64 x 64 with only two sampling steps. Code is available at https://github.com/sony/vct.
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。