AI Paper Atlas

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

Causality-inspired Federated Learning for Dynamic Spatio-Temporal Graphs

Yuxuan Liu, Wenchao Xu, Haozhao Wang, Zhiming He, Zhaofeng Shi, Chongyang Xu, Peichao Wang, Boyuan Zhang

AAAI 2026 · Federated learning and privacy › Personalized and heterogeneous FL · approach: Federated semi-supervised and graph learning · 1 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
use SC-FSGL instead of FUELSfederated STG forecasting · Average over 5 clients, 60 minMAE: 4.51 -> 4.21Improved
use SC-FSGL instead of FedAvgfederated STG forecasting · PEMSD4MAE: 3.41 -> 2.46Improved
use SC-FSGL instead of FUELSfederated STG forecasting · PEMSBAYMAE: 2.34 -> 6.44Hurt
use SC-FSGL instead of Local trainingfederated STG forecasting · PEMSBAYMAE: 2.39 -> 6.44Hurt
use SC-FSGL instead of Local trainingfederated STG forecasting · PEMSD7(M)MAE: 7.11 -> 3.50Improved
use SC-FSGL instead of GMANfederated STG forecasting · Average, 30 min horizonMAE: 3.92 -> 3.67Improved
use SC-FSGL instead of GMANfederated STG forecasting · Average, 15 min horizonMAE: 3.64 -> 3.29Improved
use SC-FSGL instead of best baseline on average over clientsfederated STG forecasting · Average, 60 minMAPE: 10.18% -> 10.08%Improved
remove causal codebook (ablation)SC-FSGL · Average over 5 datasetsMAEHurt
remove conditional separation module (ablation)SC-FSGL · Average over 5 datasetsMAEHurt

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