[2604.02183] TRU: Targeted Reverse Update for Efficient Multimodal Recommendation Unlearning

Computer Science > Artificial Intelligence arXiv:2604.02183 (cs) [Submitted on 2 Apr 2026 (v1), last revised 10 Apr 2026 (this version, v2)] Title:TRU: Targeted Reverse Update for Efficient Multimodal Recommendation Unlearning Authors:Zhanting Zhou, KaHou Tam, Ziqiang Zheng, Zeyu Ma, Yang Yang View a PDF of the paper titled TRU: Targeted Reverse Update for Efficient Multimodal Recommendation Unlearning, by Zhanting Zhou and KaHou Tam and Ziqiang Zheng and Zeyu Ma and Yang Yang View PDF HTML (experimental) Abstract: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. Approximate machine unlearning offers an efficient alternative to full retraining, yet existing methods for MRS mainly rely on a largely uniform reverse update across the model. We show that this assumption is fundamentally mismatched to modern MRS: deleted-data influence is not uniformly distributed, but concentrated unevenly across \textit{ranking behavior}, \textit{modality branches}, and \textit{network layers}. This non-uniformity gives rise to three bottlenecks in MRS unlearning: target-item persistence in the collaborative graph, modality imbalance across feature branches, and layer-wise sensitivity in the parameter space. To address this mismatch, we propose \textbf{targeted reverse update} (TRU), a plug-and-play unlearning framework for MRS. Instead of applying a blind global r...