Papers for
recommendation engineers
Papers whose findings have a practical use for this group, as judged from the abstract. Open a paper to read what it means in practice.
User activity can create false item similarity in recommender systems
Do Similarity Measures Measure Similarity? How User Activity Creates Phantom Correlation in Recommender Systems
Abstract: Item-item similarity is one of the oldest and most foundational signals in recommender systems: two items are assumed to be similar when the same users consume them. Despite the sparsity and scale of interaction data, the implicit assumption across the field is that item co-occurrence reflects genuine shared preference, capturing what users like together. In this paper, we demonstrate that a substantial fraction of observed item-item similarity is an artifact of variance in user activity rather than shared taste. Even if every user consumed items completely independently, differing only in their total consumption volume, standard similarity measures still detect strong associations between items. We term this phenomenon phantom correlation: spurious item association that emerges strictly from heterogeneous user activity in the complete absence of preference structure. Starting from minimal formal assumptions, we show that independent consumption under varying activity is described by a Rasch model, and we derive how the resulting association depends on item popularity for each similarity measure. By comparing our theoretical baseline against real-world benchmark datasets, we reveal the extent to which popular similarity measures reflect phantom correlation, which bears on how item co-occurrence is interpreted and used in collaborative filtering.
New training method improves click prediction accuracy with one network
FLMNet: Self-Ensemble Distillation from Masks to Seeds for Single-Tower CTR Prediction
Abstract: Click-through rate (CTR) prediction underpins recommender systems and online advertising, where high-dimensional categorical features interact in combinatorially many ways. Several recent CTR models, such as FinalNet and QNN, report higher accuracy when a network also learns from its own averaged predictions, for example by running the same network under different dropout patterns and training each run toward the average. Why this self-distillation helps, and whether its random sampling is needed, remains unclear. This work shows that as the number of dropout patterns grows without bound, the objective reduces exactly to the label loss of the logit-averaged prediction plus a penalty on how much the dropout copies of the network disagree. When dropout does not affect how a network gates its features, this penalty can be approximated in closed form, and FLMNet is designed on this principle and trains without sampling. Networks trained from different seeds disagree in the same form, and distilling several of them into one network recovers part of their ensemble gain. On five benchmarks against seventeen baselines, the resulting models take the top two places in accuracy with a single network at inference, and the gains of the penalty and of distillation, where measurable, largely add up.
Language embeddings can leak item and user identities in recommenders
Encoded but Exposed: Language Embeddings can Reveal More Than They Should in Recommender Systems
Abstract: Language embeddings are increasingly used in recommenders to represent item content, but their privacy risks remain poorly understood. We study whether item embeddings can reveal item identities when an adversary has access to auxiliary item text from external sources. We show that a simple cosine-similarity attack can recover item identities with high mean reciprocal rank across language models and candidate-set settings on two real-world datasets. Leakage is strongest when the adversary's partial information overlaps with the victim representation, but it also persists through semantic alignment between independently sourced descriptions. We further show that item leakage can propagate to user re-identification from sparse ratings. Finally, we evaluate obfuscated descriptions as a mitigation and find a clear leakage-utility tradeoff.
Recommendation system improves cold-start item suggestions using dual-branch graph learning
Bridging the Topological Divide: Dual-Branch Graph Distillation with Semantic Priors for Robust Cold-Start Recommendation
Abstract: Graph Collaborative Filtering models have achieved remarkable success in recommendation systems. However, they inherently suffer from a catastrophic representation collapse in strict cold-start scenarios due to the topological isolation of unobserved items. While integrating multi-modal semantic features offers a potential remedy, directly mapping isolated semantic attributes into a globally topology-driven collaborative space frequently precipitates representation degradation and optimization conflicts. To bridge this fundamental structural divide, we propose DualGCN, an innovative Dual-Branch Graph Knowledge Distillation Framework. Operating under a decoupled two-stage training paradigm, DualGCN first constructs a robust collaborative teacher network where item representations are anchored by a topology-smoothed semantic prior, effectively reshaping the latent manifold. Subsequently, a semantic student branch employs a residual mapping mechanism and a magnitude-agnostic cosine distillation objective to achieve high-fidelity cross-modal alignment without disrupting established collaborative signals. Furthermore, we introduce a dynamic norm reconstruction mechanism to rigorously calibrate the physical scales of novel and warm items during global retrieval. Extensive experiments across three real-world benchmark datasets demonstrate that DualGCN significantly outperforms state-of-the-art baselines in strict cold-start scenarios while comprehensively preserving recommendation accuracy for highly active items.
Embedding subspaces enable flexible multi-goal recommendations at scale
Embedding Subspace Partitioning for Dynamic Multi-Objective Retrieval
Abstract: Modern industrial recommender systems must optimize across competing objectives, balancing semantic relevance with business metrics such as engagement and revenue. While bi-encoders dominate large-scale retrieval due to their efficiency, they collapse these heterogeneous signals into a single static embedding space. This design creates a fundamental limitation: once trained, the retriever cannot adapt to shifting objective priorities at serving time without retraining. Moreover, joint optimization with multi-objective losses often induces interference between objectives, leading to suboptimal trade-offs. We propose Embedding Subspace Partitioning (ESP), a retrieval framework that decomposes the embedding into task-aware subspaces and replaces the single dot product with a weighted sum of per-subspace similarities, whose weights are tunable at serving time. For Transformer bi-encoders, ESP uses the model's native end-of-sequence token as a segment delimiter, with segment-aware attention masking and position encoding resets to guarantee subspace isolation in a single forward pass. Serving is performed via GPU-accelerated exhaustive kNN over one concatenated index, eliminating the need for per-objective Approximate Nearest Neighbor (ANN) infrastructure required by multi-head approaches. We evaluate ESP on an open-source benchmark built from MS MARCO. A single ESP model traces a broad Pareto frontier, consistently outperforming strong multi-task baselines across diverse operating points. In LinkedIn's job matching platform (70M+ weekly users), ESP enabled dynamic retrieval reconfiguration and delivered significant key business metric lifts.
Multi task learning improves recommendation accuracy and satisfaction
Learned Cross-Task Relationships in Multi-Task Models
Abstract: We propose a framework that learns cross-task relationships in multi-task models by approximating the joint distribution of task labels through targeted pairwise relationships. This approach improves performance via transfer learning and enhances information extraction without the intractable complexity of modeling the full joint space. Although our framework applies to any multi-task system, we demonstrate its efficacy within YouTube's production recommendation systems. Experiments across the Notifications, Homepage, and Watch Next surfaces show improvements in both accuracy and user satisfaction metrics. Finally, we propose a workflow template to facilitate broader future implementation.
Dataset recommender system helps choose evaluation data for recommender algorithms
FINALLY: A Dataset Recommender System for Recommender-Systems Research
Abstract: Dataset selection shapes the empirical conditions under which recommender-system algorithms are evaluated, yet existing tools provide limited support for constructing complete dataset sets that jointly satisfy experimental constraints and set-level selection objectives. To address this problem, I developed FINALLY, a web-based dataset recommender for constructing configurable dataset sets for offline recommender-systems evaluations. FINALLY combines required datasets, candidate-pool restrictions, metadata filters, configurable target-set sizes, Random selection, and diverse and non-diverse strategies based on adapted Effective Covariance and Convex Hull objectives. I evaluated FINALLY through 420 recommendation runs across ten systematically varied configurations. All evaluated dataset sets satisfied the applicable target-size, duplicate-avoidance, snapshot-membership, required-dataset, and metadata-filter requirements. All 40 deterministic strategy--configuration combinations were reproducible. Both the Effective-Covariance-based and Convex-Hull-based strategies produced the expected diverse-versus-non-diverse score ordering in all ten configurations. Under their corresponding objectives, the diverse strategies produced scores above all 30 configuration-specific Random results, whereas the non-diverse strategies produced scores below all 30 Random results. These results establish technical consistency for the evaluated FINALLY workflow and show that the implemented strategies follow their intended optimization directions within the investigated configuration space. They do not establish the scientific suitability, global optimality, or practical superiority of the generated selections.