AI Paper Atlas

Research map / Label-efficient and robust learning / Domain adaptation and generalization

CELL: A Causal Perspective for Fairness-aware Graph Adaptation

Hourun Li, Yifan Wang, Qinghua Ran, Junyu Luo, Jia Yang, Changling Zhou, Zhiping Xiao, Wei Ju, Xiao Luo, Ming Zhang

ICML 2026 · Label-efficient and robust learning › Domain adaptation and generalization · approach: Mixture and distillation for domain shift

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 MI-based GNN encoder (vs vanilla GCN) (ablation)CELL · syn-tΔEO: 7.65 -> 0.42Improved
add MI-based GNN encoder (vs vanilla GCN) (ablation)CELL · Pokec-nΔEO and ACC: ΔEO 0.26 -> 0.71; ACC 69.79 -> 68.93Mixed
add bipartite-aware domain alignment (ablation)CELL · syn-tΔEO: 1.47 -> 0.42Improved
add bipartite-aware domain alignment (ablation)CELL · Pokec-nΔDP: 0.79 -> 0.37Improved
add sensitive consistency learning (ablation)CELL · syn-tΔEO: 2.22 -> 0.42Improved
add sensitive consistency learning (ablation)CELL · Pokec-nΔEO: 1.17 -> 0.71Improved
add pseudo-label supervision (ablation)CELL · syn-tΔEO: 7.25 -> 0.42Improved
add pseudo-label supervision (ablation)CELL · Pokec-nΔDP: 1.22 -> 0.37Improved
add pseudo-label supervision (ablation)CELL · syn-tACC: 81.56 -> 79.70Hurt
raise pseudo-label thresholds tau, delta to at least 0.75 (ablation)CELL · PokecDP/EOImproved
increase retrieval top-K from source domain (ablation)CELL · Bail, German, Pokec, synutility and fairnessMixed
use CELL instead of GCNCELL · German-tΔDP: 15.04 -> 1.83Improved

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