2 papers
cs.AI2026
TRU: Targeted Reverse Update for Efficient Multimodal Recommendation Unlearning
Zhanting Zhou, KaHou Tam, Zeyu Ma +2
Multimodal recommendation systems (MRS) jointly model user-item interaction graphs and rich item content, but this tight coupling makes user data difficult to remove once learned.…
cs.LG2025
Dissecting Federated-Graph Aggregation under Domain Shift: Importance-Aware Aggregation via Empirical Analysis
Zhanting Zhou, Kahou Tam, Zeyu Ma +1
Federated graph learning (FGL) trains a shared graph model across clients whose local graphs differ in node features, labels, and connectivity while keeping raw graph data decentra…