以细粒度条件与语义保持采样引导扩散模型用于单样本联邦学习
Guiding Diffusion Models with Fine-Grained Conditions and Semantics-Preserving Sampling for One-Shot Federated Learning
把扩散模型当作客户端本地的数据生成器：用细粒度条件嵌入刻画类内模式、再对潜噪声分布建模并做自反思采样来提升合成数据的保真度与多样性，这套「条件嵌入+潜噪声分布采样」的组合可直接迁移到任何需要少量样本生成多样化训练数据的视觉场景（如小样本分类、长尾数据增广）。
原文依据 learns fine-grained conditional embeddings to improve semantic fidelity and diversity by modeling intra-category patterns
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论单样本联邦学习（OSFL）已成为缓解传统联邦学习高通信开销的一种有前景的范式。然而，其有效性常受客户端间数据异构性的制约。尽管近期方法利用预训练扩散模型为 OSFL 生成数据，但它们往往受制于若干实际局限，包括在捕捉本地数据细粒度特征时缺乏语义保真度，以及生成数据多样性不足，这些问题共同降低了全局模型的性能。为应对上述挑战，我们提出 Espresso，一个同时提升 OSFL 中合成数据保真度与多样性的新框架。Espresso 包含两个主要组件：（1）细粒度条件学习，通过学习细粒度条件嵌入并对类内模式建模，以提升语义保真度与多样性；（2）语义保持采样，通过对潜变量噪声的分布建模并施加自反思采样策略，使生成数据多样化。在基准数据集上的大量实验表明，Espresso 能够提升合成数据的语义保真度与多样性，从而在数据异构条件下相较最优的 OSFL 方法提升全局模型的性能。
英文原文摘要 1379 字符
One-shot Federated Learning (OSFL) has emerged as a promising paradigm to mitigate the high communication overhead of traditional federated learning. However, its effectiveness is often hindered by data heterogeneity across clients. While recent methods leverage pre-trained diffusion models to generate data for OSFL, they often struggle with some practical limitations, including a lack of semantic fidelity in capturing the fine-grained characteristics of local data, and insufficient diversity in the generated data, which collectively degrade the performance of the global model. To address these challenges, we propose Espresso, a novel framework that enhances both the fidelity and diversity of synthetic data in OSFL. Espresso consists of two main components: (1) Fine-Grained Condition Learning, which learns fine-grained conditional embeddings to improve semantic fidelity and diversity by modeling intra-category patterns, and (2) Semantics-Preserving Sampling, which diversifies the generated data by modeling the distribution of latent noises and applying a self-reflection sampling strategy. Extensive experiments on benchmark datasets demonstrate that Espresso can improve the semantic fidelity and diversity of the synthetic data, leading to an enhancement in the performance of the global model compared to state-of-the-art OSFL methods under data heterogeneity.