Jul 2026· Annual International ACM SIGIR Conference on Research and Development in Information Retrieval· pp. 812-822· 0 citations· 45 references
Computer Science
TL;DR
Because SetCDR operates directly on sets of user history records, it provides a natural way to examine how histories influence user representations, and adapts immediately to new interactions without additional retraining, enabling on-the-fly performance improvement.
Abstract
Cross-domain recommendation is a well-known technique for improving recommendations in a target domain, especially under sparse data or cold-start conditions. A common strategy is to train user embeddings separately in the source and target domains and learn a transfer function between them. In contrast, we propose SetCDR, which constructs more effective user representations in the target domain by directly incorporating each user's source and target history. We additionally introduce a lightweight domain indicator that preserves data–domain relational information. These histories, composed of item–rating pairs, are represented as variable-length sets and processed using a permutation-invariant neural architecture. This differs from conventional neural networks, which do not naturally handle unordered inputs, and is well suited to recommender systems where user history sizes vary greatly among users. The use of a permutation-invariant architecture ensures consistent embeddings regardless of input order, improving robustness to real-world variability and training efficiency. We demonstrate SetCDR in two forms: a simple sum-pooling method and an extended multihead attention–based method that captures more complex dependencies within user histories. Moreover, because SetCDR operates directly on sets of user history records, it provides a natural way to examine how histories influence user representations. Finally, SetCDR adapts immediately to new interactions without additional retraining, enabling on-the-fly performance improvement. Experimental results across multiple cross-domain benchmarks confirm that SetCDR consistently outperforms strong baselines in recommendation quality.
Cross-domain sequential recommendation (CDSR) improves target-domain prediction by leveraging multi-domain interaction histories. Most CDSR methods rely on shared entities or co-occurrence signals, which become unreliable when overlap is limited, and atomic ID representations further generalize poorly to long-tail or unseen items as cross-domain distribution shifts exacerbate this problem. Recent generative CDSR methods enable cross-domain transfer without relying on raw ID alignment by generating content-grounded semantic IDs (SIDs) for cross-domain alignment. However, two challenges remain, including (i) user-item asymmetry, with items discretized for generation whereas user preferences are encoded only implicitly in sequence representations, limiting semantic-level preference control; and (ii) selective transfer, making it difficult to assess source-domain signals against the target preference representation without an explicit discrete user anchor aligned with item IDs, which can lead to unintended transfer of irrelevant signals. This work presents UniGCRec, which constructs a cross-domain user profile from multi-domain histories and quantizes both users and items into CSC-IDs that integrate semantic and collaborative signals. This symmetric quantization places user and item representations in the same discrete CSC-ID space, effectively mitigating user-item asymmetry and enabling preference-aware selective transfer under low-overlap settings. The generator is conditioned on a user CSC-ID prefix and the target domain item CSC-ID history for next-item generation, with trie-constrained decoding ensuring target domain validity. Experiments on public multi-domain benchmarks show consistent gains over strong baselines, with particularly strong gains on several target domains.
Chaoyue Ding, Jiahao Liu, Dongsheng Li et al.· Proceedings of the 32nd ACM...· 0 citations
Large Language Models (LLMs) have demonstrated remarkable general-purpose abilities across a wide range of domains, and these strengths have also been increasingly evidenced in recommender systems. However, existing methods that attempt to integrate collaborative signals into LLMs often fail to preserve their foundational knowledge. This loss is critical in text-rich recommendation, where robust semantic understanding is required to interpret user reviews and item profiles. We propose PALRec, a parameter-preserving augmentation framework that equips an LLM with recommendation capabilities while keeping its original parameters fixed. We first construct evidence-grounded user and item profiles from reviews and use them as concise pseudo-labels for reconstruction. We then introduce lightweight, trainable user and item embedding modules optimized with a multi-task objective that combines next-item prediction and profile reconstruction. These modules are trained jointly to align collaborative signals with the LLM’s semantic space without modifying the backbone. We also employ token-aware loss decomposition and frequency-aware reweighting to stabilize training and mitigate popularity bias. Experiments on public benchmarks show that PALRec consistently outperforms fully fine-tuned counterparts in recommendation accuracy while preserving the LLM’s pre-trained knowledge. This result highlights that maintaining the LLM’s semantic understanding is crucial for effectively exploiting textual information in recommender systems.
Hyunsoo Na, Minseok Gang, Sang-goo Lee et al.· ACM Transactions on Informat...· 0 citations
The rapid growth of online digital platforms has significantly increased the need for recommender systems (RSs) that can deliver personalized content to users. Cross-domain recommender systems (CDRS) have emerged as promising solution to the limitations of single-domain models by incorporating user preferences, interaction histories, and item features from a source domain to enhance recommendations accuracy in a sparse target domain. However, effective transfer of knowledge from source domain to the target domain remains a challenging task due to differences in distributions of data, domain inconsistencies, and variations in user behavior. In this study, we propose a sparsity-aware generative adversarial networks-based cross-domain recommender system, named SPARGAN. The proposed model facilitates flexible and effective knowledge transfer by learning domain-invariant latent representations and generating realistic synthetic user-item interactions. SPARGAN incorporates adversarial learning and a domain-confusion loss to align user-item feature distributions between the source and target domains while preserving personalized user preferences. Additionally, the generator enhances the target-domain data by producing high-quality synthetic samples, thereby mitigating the impact of data sparsity problems. Extensive experiments are conducted on four real-world datasets: MovieLens, Amazon, Yelp, and Book-crossing. The experimental results demonstrate that SPARGAN consistently outperforms baseline methods in both top-N recommendation and rating prediction tasks, achieving superior performance in terms of Recall, Precision, RMSE, and F1-score under extreme sparsity conditions. Overall, this study highlights the effectiveness of adversarial learning for cross-domain knowledge transfer and provides foundation for future research on multi-source domain adaptation in cross-domain recommender systems with Gen AI models.
Matthew O. Ayemowa, Roliana Ibrahim, Noor Hidayah Zakaria et al.· Discover Computing· 0 citations
The goal of Cross-domain Recommender System (CDRS) is to recommend items in a target domain for users who have no target-domain interactions by leveraging their source-domain interaction histories. Most existing CDRSs transfer a user embedding from the source domain to the target domain and predict ratings via embedding matching with target-domain item embeddings, which can overlook fine-grained user--item preference signals expressed in reviews. To capture such fine-grained signals for each target-domain user--item pair, we propose øurs, a conditional-diffusion-based CDRS that generates a target-domain review embedding and then predicts the corresponding rating from the generated embedding. øurs~ improves review-embedding generation with two key ideas: (Idea 1) target-aware source attention to construct review guidance (\ie, a conditioning embedding); and (Idea 2) pretraining on target-domain review embeddings from target-only users to learn a broader target-domain review-embedding distribution. Extensive experiments on three cross-domain scenarios show that øurs~ consistently outperforms nine competitors, reducing MAE by up to 14.2%.
Jiwon Son, Y. Kwon, Sang-Wook Kim· Proceedings of the 32nd ACM...· 0 citations
Transformer-based sequential recommendation models, which process sequences of user-item interactions, rely heavily on the item embedding strategy. Existing approaches either use pretrained item embeddings or learn them end-to-end with the transformer. To the best of our knowledge, no prior work has compared these options from both cost and quality perspectives in a large-scale industrial setting. This paper is a case study that compares pretrained industrial graph neural network item embeddings with end-to-end trainable item embeddings across two mature production recommendation systems at Yandex: Yandex Market and Yandex Music. We additionally evaluate both approaches on a low-resource dataset sampled from Yandex Lavka production logs, for which both the data and code are publicly available for demonstration purposes. Our results show that a separate pretraining stage helps when training data is limited, but provides no worthwhile benefit for large-scale models trained on extensive datasets.
Sergei Makeev, Artem Matveev, Vladimir Baikalov et al.· arXiv.org· 0 citations
Recommender systems remain domain-bound: a model trained on one interaction environment typically requires retraining or target-domain adaptation before it can operate on a new catalogue. A recommender trained on movies cannot be directly deployed to recommend groceries or video games. Existing approaches mitigate this by transferring restricted forms of recommendation knowledge, adapting to the target domain, or leveraging large language models (LLMs) for transferable representations. We instead ask whether recommendation-specific knowledge learned solely from multiple heterogeneous domains can generalize to entirely unseen domains without target-domain adaptation or language-model pretraining. We introduce ATLAS, a multi-source recommendation domain generalization framework that learns a shared, domain-invariant user-item representation from disjoint source domains, enabling zero-shot recommendation on unseen domains. ATLAS combines a Gromov-Wasserstein alignment that preserves how users relate to one another across domains, an adversarial objective that makes item representations indistinguishable across domains, and residual vector quantization (RVQ) codebooks that compress user and item embeddings into a discrete latent space, capturing hierarchical interaction patterns while suppressing domain-specific variation. Trained on five Amazon domains and applied directly to ten unseen domains, ATLAS outperforms state-of-the-art sequential, graph-based, cross-domain, quantization-based, and LLM-based baselines on most unseen domains, with an average relative gain in HitRate of 24%. Ablations and representation analyses validate each component, and we identify a pronounced source-domain diversity effect: increasing source heterogeneity substantially improves zero-shot transfer. ATLAS establishes recommendation domain generalization as a promising paradigm for zero-shot recommendation.
Pervez Shaik, Prosenjit Biswas, A. Thorat et al.· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
