HeteroFL-LoRA is proposed, a federated framework that enables LoRA fine-tuning across heterogeneous LFMs and introduces a Singular Matrix–Guided Subspace Projection that employs singular matrices to achieve cross-subspace mapping, enabling heterogeneous clients to aggregate their LoRA updates in a unified representatio...
Zhuojia Wu, Qi Zhang, Xue-Rong Zhao et al.· Proceedings of the 32nd ACM...· 0 citations
The proposed DUE framework can effectively nullify the recall of sensitive information after MU, and even improve the performance of ULMs, and establishes a new fundamental research direction in safe training against MU vulnerabilities.
Qing Cao, Liang Hu, Dora D. Liu et al.· Proceedings of the Thirty-Fi...· 0 citations
HeteroFL-LoRA is proposed, a federated framework that enables LoRA fine-tuning across heterogeneous LFMs and introduces a Singular Matrix–Guided Subspace Projection that employs singular matrices to achieve cross-subspace mapping, enabling heterogeneous clients to aggregate their LoRA updates in a unified representatio...
Zhuojia Wu, Qi Zhang, Xuerong Zhao et al.· Proceedings of the 32nd ACM...· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
