AI Paper Atlas

Research map / Federated learning and privacy / Personalized and heterogeneous FL

Bi-level Personalization for Federated Foundation Models: A Task-vector Aggregation Approach

Yiyuan Yang, Guodong Long, Qinghua Lu, Liming Zhu, Jing Jiang

AAAI 2026 · Federated learning and privacy › Personalized and heterogeneous FL · approach: Non-IID federated learning · 2 citations (OpenAlex)

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 changedWhereMeasuredEffect
add task-vector-similarity personalized aggregation (FedBip) over uniform FedIT aggregationLLaMA-7B LoRA federated fine-tuning · Flan 8-task averageaverage score: 79.05 -> 81.21Improved
add layer-wise task-vector aggregation (FedBip-L)FedBip · Flan 8-task averageaverage score: 81.21 -> 81.90Improved
add layer-wise task-vector aggregation (FedBip-L)FedBip · OfficeHome averageaccuracy: 93.41 -> 93.74Improved
plug FedBip into FedProxFedProx · OfficeHome averageaccuracy: 89.68 -> 94.44Improved
plug FedBip into FFA-LoRAFFA-LoRA · Flan 8-task averageaverage score: 67.78 -> 73.77Improved
plug FedBip into FedDPAFedDPA · Flan 8-task averageaverage score: 81.74 -> 82.01Improved
plug FedBip into DittoDitto · OfficeHome averageaccuracy: 92.39 -> 93.63Improved
use cosine similarity instead of L2 for task-vector weights (ablation)FedBip · Flan 8-task averageaverage score: 79.55 -> 81.21Improved
use cosine similarity instead of Pearson for task-vector weights (ablation)FedBip · Flan 8-task averageaverage score: 80.77 -> 81.21Improved
compute weights from task vectors instead of parameters, and aggregate task vectors (ablation)FedBip · Flan 8-task averageaverage score: 79.63 -> 81.21Improved
aggregate task vectors rather than parameters when weights come from parameters (ablation)FedBip · Flan 8-task averageaverage score: 79.63 -> 80.15Improved
scale clients from 8 to 40 with FedBip (ablation)FedBip · Flan NLP, 40 clientsaverage score: 71.53 -> 73.00Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 3 of 3 on most 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

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.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy