# Rethinking Cross-Domain Sequential Recommendation under Open-World Assumptions

arXiv: https://arxiv.org/abs/2311.04590 · HTML: https://mlsanity.com/paper/2311.04590/

## Key facts

- Cited by 20 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 8 Nov 2023.

## Built on by (most cited first, 20 of 20)

- [A-MEM: Agentic Memory for LLM Agents](https://mlsanity.com/paper/2502.12110/) (2025)
- [OmniRouter: Budget and Performance Controllable Multi-LLM Routing](https://mlsanity.com/paper/2502.20576/) (2025)
- [PersonaX: A Recommendation Agent Oriented User Modeling Framework for Long Behavior Sequence](https://mlsanity.com/paper/2503.02398/) (2025)
- [SLMRec: Distilling Large Language Models into Small for Sequential Recommendation](https://mlsanity.com/paper/2405.17890/) (2024)
- [Bridge the Domains: Large Language Models Enhanced Cross-domain Sequential Recommendation](https://mlsanity.com/paper/2504.18383/) (2025)
- [LLM-RecG: A Semantic Bias-Aware Framework for Zero-Shot Sequential Recommendation](https://mlsanity.com/paper/2501.19232/) (2025)
- [Fine-Grained Dynamic Framework for Bias-Variance Joint Optimization on Data Missing Not at Random](https://mlsanity.com/paper/2405.15403/) (2024)
- [Knowledge Graph Pruning for Recommendation](https://mlsanity.com/paper/2405.11531/) (2024)
- [Beyond KAN: Introducing KarSein for Adaptive High-Order Feature Interaction Modeling in CTR Prediction](https://mlsanity.com/paper/2408.08713/) (2024)
- [Token-Controlled Re-ranking for Sequential Recommendation via LLMs](https://mlsanity.com/paper/2511.17913/) (2025)
- [LLM-EDT: Large Language Model Enhanced Cross-domain Sequential Recommendation with Dual-phase Training](https://mlsanity.com/paper/2511.19931/) (2025)
- [Generative Archetype-Grounded Item Representations for Sequential Recommendation](https://mlsanity.com/paper/2606.11023/) (2026)
- [Sharpness-aware Model Merging with Salience Recovery for LLM-based Cross-Domain Sequential Recommendation](https://mlsanity.com/paper/2607.25366/) (2026)
- [Selection and Exploitation of High-Quality Knowledge from Large Language Models for Recommendation](https://mlsanity.com/paper/2508.07223/) (2025)
- [Debiasing Sequential Recommendation with Time-aware Inverse Propensity Scoring](https://mlsanity.com/paper/2603.04986/) (2026)
- [Social Knowledge for Cross-Domain User Preference Modeling](https://mlsanity.com/paper/2603.10148/) (2026)
- [MOSAIC: Multi-Domain Orthogonal Session Adaptive Intent Capture for Prescient Recommendations](https://mlsanity.com/paper/2604.10147/) (2026)
- [Breaking the Information Silo: Semantic Personas for Cross-Domain Recommendation](https://mlsanity.com/paper/2606.01783/) (2026)
- [Empowering Cross-Domain Sequential Recommendation with Hybrid Tokenization and Serial-Parallel Decoding](https://mlsanity.com/paper/2607.28659/) (2026)
- [CoRCi: Cross-Reconstruction of Coherent Interests Modeling in Cross-Domain Sequential Recommendation](https://mlsanity.com/paper/2608.09580/) (2026)

## Abstract (arXiv)

Cross-Domain Sequential Recommendation (CDSR) methods aim to tackle the data sparsity and cold-start problems present in Single-Domain Sequential Recommendation (SDSR). Existing CDSR works design their elaborate structures relying on overlapping users to propagate the cross-domain information. However, current CDSR methods make closed-world assumptions, assuming fully overlapping users across multiple domains and that the data distribution remains unchanged from the training environment to the test environment. As a result, these methods typically result in lower performance on online real-world platforms due to the data distribution shifts. To address these challenges under open-world assumptions, we design an Adaptive Multi-Interest Debiasing framework for cross-domain sequential recommendation (AMID), which consists of a multi-interest information module (MIM) and a doubly robust estimator (DRE). Our framework is adaptive for open-world environments and can improve the model of most off-the-shelf single-domain sequential backbone models for CDSR. Our MIM establishes interest groups that consider both overlapping and non-overlapping users, allowing us to effectively explore user intent and explicit interest. To alleviate biases across multiple domains, we developed the DRE for the CDSR methods. We also provide a theoretical analysis that demonstrates the superiority of our proposed estimator in terms of bias and tail bound, compared to the IPS estimator used in previous work.