Research map / Federated learning and privacy / Personalized and heterogeneous FL
Decoupled Low-Rank Adaptation for Robust Federated Fine-Tuning
Xiuwen Fang, Xuliang Yang, Mang Ye
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add decoupled LoRA fine-tuning (dual-branch sample partition) after warm-up (ablation) | FedDR-LoRA · CIFAR-100, Pairflip 0.2 | accuracy (%): 88.08 -> 90.54 | Improved |
| add decoupled LoRA fine-tuning after warm-up (ablation) | FedDR-LoRA · CIFAR-100, Pairflip 0.4 | accuracy (%): 85.41 -> 87.37 | Improved |
| add noisy branch negative learning on top of DLF (ablation) | FedDR-LoRA · CIFAR-100, Pairflip 0.4 | accuracy (%): 87.37 -> 88.30 | Improved |
| add noisy branch negative learning on top of DLF (ablation) | FedDR-LoRA · CIFAR-100, Symflip 0.2 | accuracy (%): 90.66 -> 91.37 | Improved |
| add DivideMix to FedLR | FedLR · CIFAR-100, Pairflip 0.4 | accuracy (%): 60.11 -> 70.12 | Improved |
| add Relabel to FedLR | FedLR · CIFAR-100, Symflip 0.2 | accuracy (%): 85.42 -> 75.31 | Hurt |
| add SCE loss to FedLR | FedLR · CIFAR-100, Pairflip 0.4 | accuracy (%): 60.11 -> 66.79 | Improved |
| aggregate only B (clean and noisy separately) and keep A local, as in FedDR-LoRA, versus aggregating A and B | FedDR-LoRA · CIFAR-100 | accuracy (%) | Improved |
| use FedDR-LoRA instead of baselines under instance-dependent noise 0.4 | FedDR-LoRA · CIFAR-100, instance-dependent noise | accuracy (%): 72.44 -> 88.23 | Improved |
| use FedDR-LoRA under heterogeneous per-client noise rates | FedDR-LoRA · CIFAR-100, Pairflip heterogeneous | accuracy (%): 77.07 -> 88.33 | Improved |
| FedDR-LoRA on a text task (RoBERTa-base) with low noise Symflip 0.2 | FedDR-LoRA · 20 Newsgroups, Symflip 0.2 | accuracy (%): 66.42 -> 66.53 | Mixed |
Closest papers by meaning
- Robust Federated Finetuning of LLMs via Alternating Optimization of LoRA (NeurIPS 2025 · similarity 0.91)
- FedSVD: Adaptive Orthogonalization for Private Federated Learning with LoRA (NeurIPS 2025 · similarity 0.90)
- Adaptive LoRA Experts Allocation and Selection for Federated Fine-Tuning (NeurIPS 2025 · similarity 0.90)
- Fed-SB: A Silver Bullet for Extreme Communication Efficiency and Performance in (Private) Federated LoRA Fine-Tuning (NeurIPS 2026 · similarity 0.89)
- Improving LoRA in Privacy-preserving Federated Learning (ICLR 2024 · similarity 0.89)
- Selective Aggregation for Low-Rank Adaptation in Federated Learning (ICLR 2025 · similarity 0.89)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
