Context-Aware Disentanglement for Cross-Domain Sequential Recommendation: A Causal View
2026-04-09 • Information Retrieval
Information Retrieval
AI summaryⓘ
The authors study how to improve recommendations across different domains by sharing knowledge, but they note current methods have problems like confusing context effects and conflicts when learning from multiple domains. They propose a new method called CoDiS that carefully separates what preferences are shared across domains from what is domain-specific, using techniques to reduce misleading context and avoid learning conflicts. Their tests on real datasets show their method works better than existing approaches. This helps make recommendations more accurate even when users don't overlap much between domains.
Cross-Domain Sequential RecommendationUser PreferencesContext ConfoundingGradient ConflictDomain-Shared PreferencesDomain-Specific PreferencesVariational MethodsDisentanglementCold-Start Problem
Authors
Xingzi Wang, Qingtian Bian, Hui Fang
Abstract
Cross-Domain Sequential Recommendation (CDSR) aims to en-hance recommendation quality by transferring knowledge across domains, offering effective solutions to data sparsity and cold-start issues. However, existing methods face three major limitations: (1) they overlook varying contexts in user interaction sequences, resulting in spurious correlations that obscure the true causal relationships driving user preferences; (2) the learning of domain- shared and domain-specific preferences is hindered by gradient conflicts between domains, leading to a seesaw effect where performance in one domain improves at the expense of the other; (3) most methods rely on the unrealistic assumption of substantial user overlap across domains. To address these issues, we propose CoDiS, a context-aware disentanglement framework grounded in a causal view to accurately disentangle domain-shared and domain-specific preferences. Specifically, Our approach includes a variational context adjustment method to reduce confounding effects of contexts, expert isolation and selection strategies to resolve gradient conflict, and a variational adversarial disentangling module for the thorough disentanglement of domain-shared and domain-specific representations. Extensive experiments on three real-world datasets demonstrate that CoDiS consistently outperforms state-of-the-art CDSR baselines with statistical significance. Code is available at:https://anonymous.4open.science/r/CoDiS-6FA0.