Research map / Federated learning and privacy / Personalized and heterogeneous FL
FedTreeLoRA: Reconciling Statistical and Functional Heterogeneity in Federated LoRA Fine-Tuning
Jieming Bian, Lei Wang, Letian Zhang, Jie Xu
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 layer-wise adaptive cluster count (vs fixed k=4) (ablation) | FedTreeLoRA · GLUE (non-IID alpha=0.5, 20 clients) Average | accuracy (%): 91.45 -> 92.36 | Improved |
| add layer-wise adaptive cluster count (vs fixed k=8) (ablation) | FedTreeLoRA · GLUE (non-IID alpha=0.5, 20 clients) Average | accuracy (%): 90.74 -> 92.36 | Improved |
| add layer-wise adaptive cluster count (vs global k=1) (ablation) | FedTreeLoRA · GLUE (non-IID alpha=0.5, 20 clients) Average | accuracy (%): 87.20 -> 92.36 | Improved |
| add global tree skeleton (vs independent per-layer clustering) (ablation) | FedTreeLoRA · GLUE (non-IID alpha=0.5, 20 clients) Average | accuracy (%): 89.47 -> 92.36 | Improved |
| add external expert with scalar mixing (vs cluster-only) (ablation) | FedTreeLoRA · GLUE (non-IID alpha=0.5, 20 clients) Average | accuracy (%): 91.40 -> 92.36 | Improved |
| replace scalar mixing with decomposed experts (ablation) | FedTreeLoRA · GLUE (non-IID alpha=0.5, 20 clients) Average | accuracy (%): 92.36 -> 92.57 | Improved |
| replace scalar mixing with MoE router (ablation) | FedTreeLoRA · GLUE (non-IID alpha=0.5, 20 clients) Average | accuracy (%): 92.36 -> 92.02 | Hurt |
| cluster-only variant vs strongest baseline FedLEASE (ablation) | FedTreeLoRA · GLUE (non-IID alpha=0.5, 20 clients) Average | accuracy (%): 91.19 -> 91.40 | Improved |
| FedTreeLoRA vs FedLEASE on GLUE | FedTreeLoRA · GLUE (non-IID alpha=0.5, 20 clients) Average | accuracy (%): 91.19 -> 92.36 | Improved |
| FedTreeLoRA vs FedALT on FLAN | FedTreeLoRA · FLAN average (LLaMA-2-7B) | accuracy (%): 61.77 -> 63.19 | Improved |
| double FFA-LoRA rank from 4 to 8 | FFA-LoRA · GLUE (non-IID alpha=0.5, 20 clients) Average | accuracy (%): 86.93 -> 87.40 | No clear effect |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.
- FedALT — FedALT: Federated Fine-Tuning Through Adaptive Local Training with Rest-of-World LoRA (AAAI 2026) · better on 2 of 2 shared results
- FedSA — FedSA: A Unified Representation Learning via Semantic Anchors for Prototype-based Federated Learning (AAAI 2025) · better on 2 of 2 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- FLoRA: Federated Fine-Tuning Large Language Models with Heterogeneous Low-Rank Adaptations (NeurIPS 2024 · similarity 0.92)
- 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 0.90)
- Robust Federated Finetuning of LLMs via Alternating Optimization of LoRA (NeurIPS 2025 · similarity 0.89)
- Improving LoRA in Privacy-preserving Federated Learning (ICLR 2024 · similarity 0.89)
- Differentially Private Federated Low Rank Adaptation Beyond Fixed-Matrix (NeurIPS 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.
