Robust Cross-Domain Sequential Recommendation Driven by Causal Disentanglement and Counterfactual Generation
Sequential recommendation systems are confronted with the dual challenges of data sparsity and domain bias. Especially in cross-domain scenarios, user interests are deeply coupled with domain-specific noise, which severely restricts recommendation performance. Targeted at improving the robustness of cross-domain sequen...