Li Chen

Personalized recommendations are a ubiquitous part of online experiences, and recommender systems have become a major driver of business value across many digital platforms. Numerous success stories across a range of application domains highlight their broad impact, yet significant practical challenges remain unresolved. At the same time, although academic research in this field is thriving and has historically contributed valuable solutions, a concerning gap between academia and industry is emerging. In particular, the increasing emphasis on ever more complex machine learning models and use-case-agnostic evaluation methods raises the question of whether academic efforts are aligned with the most pressing real-world needs. In this article, we first survey successful recommender system applications from the literature, illustrating their wide applicability and the ways in which they generate business value. We then examine the contributions of academic research in relation to the practical challenges faced in industry. Finally, we issue a call to action for stronger collaboration between academia and industry. We argue that academia should focus more often on problems of practical relevance. In parallel, industry partners should more actively share open challenges and provide access to real-world datasets, enabling researchers to better address practical problems and close the existing gap.

Cross-domain recommendation (CDR) leverages source domain knowledge to alleviate data sparsity in target domains. While existing methods have achieved significant success by utilizing overlapping users as bridges for knowledge transfer, they inadvertently introduce a critical fairness issue we term CDR Overlapping Bias : recommendation quality significantly improves for overlapping users but often stagnates or degrades for non-overlapping users. Furthermore, we observe that this disparity exacerbates as the ratio of overlapping users decreases, a common scenario in real-world applications. To address this inequity, we propose a novel model-agnostic framework named S parsity- C alibrated V irtual U ser G eneration (SC-VUG). SC-VUG synthesizes virtual source profiles for non-overlapping target users, effectively transforming them into virtual overlapping users to enable equitable knowledge transfer. Specifically, we design a Sparsity-Calibrated Generator that employs a dual-view attention mechanism to model virtual users as probabilistic distributions, where the variance explicitly captures generation uncertainty related to the overlapping ratio. To ensure the reliability of these synthesized profiles, we introduce a Structure-Preserving Limiter incorporating Anchor Loss and Isometry Loss, which aligns generated users with the source distribution while preserving the topological structure of user preferences. Comprehensive experiments conducted on three public datasets with five CDR baselines demonstrate that our method effectively mitigates the CDR non-overlapping user bias, while maintaining comparable overall accuracy in most settings. Our code is publicly available at https://github.com/WeixinChen98/VUG/tree/sc-vug.

While Cross-Domain Sequential Recommendation (CDSR) has proven effective in mitigating data sparsity and enhancing accuracy, its impact on recommendation diversity remains largely unexplored. We are the first to reveal a counterintuitive phenomenon: while CDSR improves accuracy, it often comes at the cost of diversity, confining users to a narrower scope of interests. Through rigorous empirical experiments and theoretical analysis, we pinpoint two fundamental determinants driving this decline: (1) Domain Homogeneity, where excessive similarity between domains enforces preference redundancy; and (2) Information Asymmetry, where insufficient signal from the source domain fails to meaningfully perturb target-domain distributions. To address these challenges, we propose DivCDSR, a novel model-agnostic framework designed to enhance diversity in CDSR. Specifically, we introduce a dual-prototype semantic constraint mechanism that mitigates the homogenization trap via intra-domain clustering with orthogonalization and inter-domain separation. Furthermore, we devise a dual-guided diffusion module that augments source-domain information by generating new sequences, thereby resolving the information asymmetry issue. Extensive experiments on three public datasets demonstrate that our DivCDSR significantly enhances diversity metrics while maintaining or even improving recommendation accuracy simultaneously. %, offering a robust solution to the informational poverty inherent in conventional CDSR models. All datasets and code are available at https://github.com/Asuei-cs/DivCDSR.

Expressive arts therapy supports creative expression through visual art, music, dance, and drama, fostering emotional awareness, regulation, and personal growth. Recent advances in generative AI, particularly multimodal models that produce images, music, text, and soundscapes, create new opportunities to scaffold creative exploration, provide adaptive prompts, and support reflective dialogue in therapeutic contexts. At the same time, integrating AI into expressive arts therapy raises important challenges around agency, authorship, emotional alignment, ethics, data privacy, and preserving therapeutic relationships. This workshop brings together researchers, designers, therapists, and practitioners to explore how human-centered AI can support creativity in art and music therapy while respecting therapeutic values. We focus on insights from AI-enabled therapy systems, design approaches balancing creativity and emotional well-being, and evaluation methods that extend beyond usability and traditional clinical outcomes. Through interdisciplinary discussion, the workshop aims to surface open research questions and design considerations for AI use in expressive arts therapy.

Offline evaluation is the predominant method for scientific research in recommender systems, enabling the comparison of alternative recommendation approaches using pre-collected datasets and computational metrics without involving human participants. However, recent reproducibility studies reveal that many offline evaluations in the literature lack scientific rigor or adopt research practices that cast doubt on the validity of their findings. A particularly common and ultimately catastrophic flaw is the comparison of newly proposed machine learning models against untuned or poorly tuned baseline models. Combined with limited reproducibility, such practices raise serious concerns about the true progress achieved by increasingly complex recommendation algorithms. In this editorial, we argue for stronger methodological standards and summarize essential guidance and best practices for conducting rigorous offline evaluations of recommender systems. Accordingly, ACM Transactions on Recommender Systems will place increased emphasis on methodological rigor in all future submissions, with particular priority given to work that provides comprehensive reproducibility materials and clearly documents the tuning procedures used for baseline models.