🛰️ Daily AI Frontier
‹ back to 2026-07-24

Towards Privacy-Preserving Federated Prompt Tuning under Data Heterogeneity: A Subspace-Decomposed Expert Approach

Research Multimodal & Generative

Merged summary

TL;DR - FedSEPT is a privacy-preserving federated prompt-tuning method for vision-language models that represents multiple prompt experts in compact shared subspaces. It aims to improve personalization and global generalization under data heterogeneity without expanding differential-privacy noise and communication costs.

  • Shared low-rank factors, a fixed public basis, and private residuals decompose prompt experts into compact components.
  • Only the compact factor space is communicated and perturbed for local differential privacy, enabling direct server aggregation.
  • On-device routing adaptively combines complementary experts, with cached text features supporting efficient logit-level fusion.
  • Across 11 heterogeneous benchmarks, FedSEPT reportedly outperforms strong baselines on the local-versus-global trade-off under equal privacy constraints.

Sources (1)

Towards Privacy-Preserving Federated Prompt Tuning under Data Heterogeneity: A Subspace-Decomposed Expert Approach

arXiv cs.CV Yuhua Wang, Xiaodong Li, Yihao Guo, Yuxiang Jia, Qinnan Zhang, Yifan Sun, Hainan Zhang, Yongxin Tong, Zhiming Zheng 2026-07-23 arXiv:2607.21417

TL;DR - FedSEPT is a privacy-preserving federated prompt-tuning method for vision-language models that represents multiple prompt experts in compact shared subspaces. It aims to improve personalization and global generalization under data heterogeneity without expanding differential-privacy noise and communication costs.

  • Shared low-rank factors, a fixed public basis, and private residuals decompose prompt experts into compact components.
  • Only the compact factor space is communicated and perturbed for local differential privacy, enabling direct server aggregation.
  • On-device routing adaptively combines complementary experts, with cached text features supporting efficient logit-level fusion.
  • Across 11 heterogeneous benchmarks, FedSEPT reportedly outperforms strong baselines on the local-versus-global trade-off under equal privacy constraints.
item →