Papers for
streaming service developers
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.
Recommendation ranking improves with sparse feedback using pretrained user routed adapters
SAGE-PAIR: Source-Pretrained Routed Residual Adaptation for Sparse-Feedback Recommendation
Abstract: Recommendation systems often enter a new serving window before enough outcomes are available to retrain the deployed model. We study this regime with 0.05-1.0% target feedback, a range that spans our stress tests while keeping supervision far below the degrees of freedom of the source recommender. At this scale, existing fine-tuning and meta-learning methods still estimate target-specific updates from extremely sparse feedback, leaving the update direction weakly identified. SAGE-PAIR instead transfers a source-supported residual correction space and only calibrates this space using target outcomes. We introduce SAGE-PAIR, a staged adapter that pretrains a low-dimensional global residual on an adjacent source period and adds a soft router over residual experts conditioned on source user profiles. Target feedback then calibrates this residual state rather than relearning the recommender. On a locked test split of KuaiSAR-v2, SAGE-PAIR (Full) achieves a 1.77% relative improvement in NDCG@10, while a core-protocol comparison shows a further 0.25% gain from routed experts beyond source-pretrained PAIR. Matched source-pretrained LoRA and Adapter-Tuning baselines remain near the source model, supporting the need for a different adaptation space. Thus, SAGE-PAIR improves ranking quality with millisecond-scale adaptation and inference costs. Sparse-feedback adaptation is fundamentally governed by the identifiability of the update space rather than by label count alone. Source pretraining supplies estimable correction directions for scarce labels, while user-conditioned routing contributes a consistent, differentiated cross-user correction on top of the global correction. Our code is available at https://anonymous.4open.science/r/SAGE-PAIR-610E/.
New method improves generative recommendation by learning from user history
Rethinking Latent Reasoning for Generative Recommendation: Beyond Teacher-Generated CoT and Recurrent Hidden-State Feedback
Abstract: Latent reasoning in generative recommendation faces two practical barriers: constructing teacher-generated chain-of-thought (CoT) supervision is costly, and recurrent hidden-state feedback requires a specialized execution loop. We revisit both the supervision source and the computation interface. A step-level diagnosis measures how recommendation CoT supports next-item prediction and trace coherence. In the analyzed traces, history analysis is the only step with positive mean contributions on both dimensions, while later preference and intent steps have negative mean answer contributions. Guided by this finding, we propose PHiT (Prefilled, History-Informed Thought Tokens), which learns intermediate states from semantic targets extracted directly from observed item histories. PHiT aggregates cached sparse item encodings to construct these targets and supervises the contextual states of distinct vocabulary thought tokens. All thought tokens are supplied together as ordinary input IDs during a single causal prefill, removing the need for recurrent final-state feedback. With supervised fine-tuning alone, PHiT outperforms the compared baselines across four recommendation benchmarks. Controlled ablations support the value of input-matched semantic supervision, while cost measurements show fast target construction with training time comparable to the recommendation-only thought-token baseline.
Optimized semantic tokenization improves large model based recommendations
OptST: Optimizing Semantic Tokenization for LLM-Enhanced Recommendation
Abstract: Semantic information can enrich the identifiers and structured fields used by recommender systems. Semantic tokenization provides a structured representation of this information by converting LLM-derived user profiles into discrete tokens. In the common two-stage pipeline, the tokenizer is trained before the recommender. Its token assignments are then treated as fixed input features during recommendation training. As a result, the downstream recommender can learn how to use the assigned indices, but it cannot revise the semantic partition that produced them. Making this partition task-aware is difficult because nearest-code assignment is discrete, and unconstrained adaptation may also distort the geometry learned during pretraining. Moreover, a fixed token depth exposes the same semantic resolution to every interaction. We thus propose OptST (Optimizing Semantic Tokenization), which makes a pretrained tokenizer responsive to downstream recommendation. To revise token assignments without directly updating the pretrained encoder or base codebooks, OptST learns a low-rank deformation whose drift is limited by an anchor regularizer. Recommendation supervision reaches this transformation through a straight-through estimator, while prediction retains hard nearest-code assignments. Beyond adapting the assignments themselves, OptST selects an interaction-conditioned prefix to control how much residual detail is exposed to the recommender. Experiments on four public datasets show that OptST achieves the highest AUC among conventional and LLM-enhanced baselines. Its improvements also persist across different recommendation backbones. These results establish task-aware semantic tokenization as an effective interface between LLM-derived profiles and downstream recommendation models.
Transformers improve recommendation by predicting many possible next items
Compote: Collapse-Resistant Multiple Hypothesis Learning For Sequential Recommendation
Abstract: Next-item prediction is inherently ambiguous: a single interaction history often admits multiple plausible continuations across distinct catalogue regions. Standard sequential recommenders compress these alternatives into a single prediction, conflating user intents. We introduce COllapse-resistant Multiple hyPOThesis lEarning (COMPOTE), a multi-hypothesis transformer trained with an annealed objective that transitions from uniform supervision to a winner-takes-all assignment. This mechanism forces hypotheses to specialize into distinct catalogue regions, while inference relies on an efficient, normalizer-free max-similarity rule. Theoretically, we show that in the population case, the optimal solutions partition the conditional next-item distribution according to a max-entropy quantization principle, while pooled scores preserve the Bayes-optimal ranking. We also prove a generalization bound scaling logarithmically with the number of hypotheses. Across three Amazon Reviews 2023 benchmarks, COMPOTE outperforms strong ID-based, generative, and multi-interest baselines using item IDs alone. Matched-capacity ablations confirm that performance gains stem directly from the annealed objective, preventing the head collapse observed in conventional multi-head architectures, rather than added capacity, ensembling, or routing.
JevRec improves item recommendation by encoding popular items with shorter codes
JevRec: Linguistic Encoding with Jev-like Decision Modeling for LLM-based Generative Recommendation
Abstract: Generative recommendation (GR) represents items as semantic identifiers (SIDs) and autoregressively generates the next item's SID from interaction history. However, most methods assign the same SID length to all items despite highly unequal interaction frequencies. This uniform design repeatedly incurs the cost of fine-grained codes for popular items, while shortening codes to reduce this cost also compresses the semantic representations of long-tail items. Inspired by the compact forms of frequent expressions in natural language, we propose JevRec to encode items as variable-length words and generate the next item from interaction context. Linguistic encoding retains frequency-dependent semantic prefixes from shared codebooks, producing shorter SIDs for popular items and richer descriptions for long-tail items. Yet predicting a sequence-level length and generating token-level content impose different optimization requirements that may conflict under a shared output head. Our Jev-like decision modeling component therefore predicts a length distribution with a dedicated head and generates content conditioned on length through a separate head while sharing the interaction-history representation. Experiments on three public datasets show that JevRec outperforms all compared baselines across the evaluated recommendation metrics. Further analyses show lower training and inference time than TIGER and reduced gradient conflict between length and content prediction.
Item similarity measures gain theoretical foundations for recommender systems
Theoretically Sound Item Similarity Measures for Recommender Systems
Abstract: Quantifying the similarity between items is fundamental to neighborhood-based collaborative filtering, yet many commonly used similarity measures are heuristics that lack theoretical justification. The use of heuristics forces practitioners to rely on empirical trial-and-error, offering little insight into why certain measures outperform others or how they might be improved. In this paper, we present a principled approach for designing item similarity measures grounded in probabilistic modeling and parameter estimation. By adopting this formal perspective, we derive two previously overlooked similarity measures based on the Bernoulli and multinomial distributions. Through an extensive ablation study we highlight the critical role of bias, probabilistic smoothing, treating the unknowns correctly, and choosing appropriate probability distributions. Finally, we show that the commonly used cosine similarity can be interpreted as the solution to a specific optimization problem, thereby providing a novel perspective on this decades-old heuristic.
Generative recommendation improves scaling with efficient recurrent depth
LoopGR: Parameter-Efficient Scaling for Generative Recommendation via Recurrent Depth
Abstract: Generative recommendation formulates item recommendation as autoregressive generation rather than a discriminative scoring problem. Although empirical studies exhibit favorable scaling behavior in recommendation, unleashing the benefits of scaling remains challenging under the strict model size and inference cost constraints of real-world systems. A central challenge is to combine expressive modeling of user-item interaction histories with flexible computational depth without proportionally increasing model size. We introduce LoopGR, a parameter-efficient scaling framework for generative recommendation via recurrent depth. LoopGR employs a tailored hierarchical recurrent depth mechanism that alternates between a shared generative decoder and a lightweight context updater. For efficient inference, we further develop recurrent trajectory self-distillation that distills a low-budget student from a high-budget teacher by aligning both predictive distributions and evolution directions. Extensive experiments show that LoopGR consistently outperforms strong generative recommendation baselines. Compared with a representative encoder-decoder baseline, LoopGR achieves superior performance while reducing model size by up to 52%. These results establish recurrent depth as a new scaling dimension, offering a flexible and resource-efficient scaling path for generative recommendation.
Knowledge distillation improves parallel code prediction for recommendations
Knowledge Distillation across Item and Code Spaces for Parallel Generative Recommendation
Abstract: Generative recommendation with semantic IDs represents each item as a sequence of discrete codes derived from its semantic representation and predicts these codes to generate recommendations. Existing semantic-ID recommenders are typically trained with hard labels for the observed next item. Although such supervision identifies the target, it provides no explicit distributional information about alternative items conditioned on the same user history. Sequential recommenders can supply this item-level predictive knowledge, but transferring it to semantic-ID generation is challenging because the teacher and student operate in item and code spaces, respectively. We propose two complementary knowledge-distillation objectives for parallel semantic-ID recommenders, which predict all code positions simultaneously without autoregressive prefix conditioning. Item Distribution Matching (IDM) reconstructs a catalogue-level item distribution from the student's parallel code predictions and aligns it with the teacher's item distribution. Code Marginal Matching (CMM) projects the teacher's item probabilities onto each code position using fixed item-to-code assignments, then matches the resulting code marginals to the corresponding student heads. Both objectives transfer the teacher's soft distributional knowledge while preserving the student's parallel prediction structure. Experiments on benchmark datasets show that both methods consistently outperform existing state-of-the-art approaches. On average, IDM improves Recall@10 by 15.39%, while CMM improves NDCG@10 by 8.28%. These results demonstrate that distillation in either reconstructed item space or code-marginal space can substantially strengthen hard-label supervision for parallel generative recommendation.
Decision4rec improves recommendations using decision models and knowledge graphs
Decision4Rec: A Jev-Inspired Framework of Decision Models on Knowledge and Active Graphs for Multi-Agent Recommender Systems
Abstract: Decision models, such as the commercial Jev, are language models that return one probability for each option of a question instead of generating text. We present Decision4Rec, an open and modular framework that makes decision models useful for recommendation. A knowledge graph writes the logs of each recommendation scenario as short statements on a defined scale, such as "sold: top 10%". A pool of agents, proposes candidates, and a residual ranker keeps the best 30. A decision model, trained once on real choices of three scenarios, reads them through one interface of typed questions, and a fusion step combines its answers with the score of the ranker. Any model with this interface can be used, and we run the commercial Jev, the released open Kev and Kev models trained by us on recommendation requests. The framework with a trained 4B decision model is 25 to 36% above the best trained baseline in NDCG@10 in three sparse Amazon catalogues, one of them never used to train the decision model, and almost tie with baselines on the dense Steam dataset. Most of the gain comes from the pool, and with a trained recommender it comes from what the agents know about its candidates rather than from new candidates. The 27B decision model adds to the pool 5 to 6% above the 0.8B model in three of four domains. A trained 0.8B open model beats the released Jev in eleven of fifteen settings and is never worse. On MovieLens-1M, a dense domain never seen in training, the gain is 5 to 7%. Code, configurations and trained models are available at https://github.com/marcopoli/decision4rec and https://huggingface.co/m-polignano.
Generative recommendation improves item predictions with balanced semantic codes
Structure-Aligned History Learning with Subspace-Structured Semantic IDs for Generative Recommendation
Abstract: Generative recommendation constructs semantic IDs for catalog items and predicts the next ID from user interaction histories. However, existing methods face limitations in both stages. First, conventional quantizers may produce imbalanced code assignments and lack coherent intermediate semantic units between individual tokens and complete items, limiting history reconstruction. Second, recommendation models remain dominated by next-item prediction, treating historical interactions mainly as conditioning context and providing limited supervision for recovering persistent preferences across the historical path, particularly when relevant preference signals are distributed over distant interactions. To address these limitations, Structure-Aligned History Learning (SAHL) reconnects the two stages by using the structure created during quantization directly for history reconstruction. Its Semantic Residual Balanced Quantization (SRBQ) forms a balanced token-subspace-item hierarchy, and SAHL reconstructs history at these same granularities. Predictive entropy prioritizes uncertain transitions, while curriculum optimization progressively shifts training to inference-aligned next-item prediction. On four Amazon Review categories, SAHL improves NDCG@10 by 18.6%-30.0% over RPG and 2.7%-6.0% over strong method MHL. Controlled analyses verify balanced code use, the benefit of following quantizer-defined groups during reconstruction, and dependence on user history. The relevant code is available at https://github.com/CQU-MM-Intelligent-Lab/SAHL.
Language models boost personalized recommendation speed and quality
DuoLM: Fast and Effective Personalized Recommendation with Language Models
Abstract: Language models can make personalized recommendations from rich item content, user history, and feedback. However, evaluating many candidates with full-text inputs incurs substantial online computation and limits serving throughput. We introduce DuoLM, a framework for fast and effective personalized recommendation that integrates a compact recommendation model with an efficient serving runtime. An encoder LLM learns compact item representations for offline reuse, while a personalized ranker LLM combines them with user history and feedback for recommendation. We train both models end to end with recommendation supervision and teacher distillation. The runtime further accelerates inference through virtual token IDs for embedding access, shared history computation, and CUDA Graph replay. Benchmark results show that our framework achieves recommendation quality comparable to that of the expert teacher model. On a single H100, it delivers approximately 26,400 candidate scores per second, about $60\times$ the throughput of the teacher served with vLLM and automatic prefix caching.
Reinforcement learning improves recommendations by adapting choices dynamically
Reinforcement Learning with Conformal Action Sets: An Application to Sequential Recommendation
Abstract: Sequential recommenders typically use a fixed slate size even though the number of useful alternatives changes within a session. We propose Reinforcement Learning with Calibrated Pruning (RLCP), which adapts the retained action set using critic scores and an online threshold. The threshold is updated from binary feedback indicating whether the set contains an action in a proxy target. We prove a deterministic bound on the observed proxy miss rate along adaptive trajectories. To quantify the effect of pruning on reward, we derive an exact decomposition of value loss into filtering and selection losses. Under explicit proxy and critic approximation conditions, this decomposition yields a finite session reward bound that also accounts for imperfect selection and set truncation, without requiring the learning parameters to converge. Experiments on KuaiRand-Pure and MovieLens 1M compare two RLCP implementations with four RL baselines. In each of the 19 configurations, at least one RLCP variant achieves the highest catalog diversity, reaching $1.11\times$ to $5.21\times$ that of the strongest baseline, with competitive session depth and no larger retained sets.
LLM powered personal prompts improve sequential item recommendations
Learning Robust Personalized Prompts for LLM-Driven Sequential Recommendation
Abstract: LLM-driven sequential recommendation formulates next-item prediction as autoregressive generation conditioned on natural-language prompts. However, minor wording changes in semantically equivalent prompts can cause substantial performance fluctuations, undermining robustness and requiring costly manual prompt engineering. Continuous prompt learning reduces template dependence but faces two interacting challenges: shared task-level instructions lack user-specific reasoning guidance, while gradient updates can push continuous prompts outside the LLM's effective semantic space. Injecting personalized signals can further amplify this semantic drift. To address these challenges, we propose LRPRec, a learnable prompting framework that initializes continuous instruction prompts from discrete templates and introduces two complementary mechanisms. Personalized prompt injection encodes user behavior into a preference embedding and additively injects it into shared prompts, enabling parameter-efficient user-level adaptation. A semantic drift constraint regularizes the shared prompts within a trust region around their initialization anchors to preserve semantic validity during optimization. By constraining the shared component while allowing additive personalization, LRPRec decouples stability from expressiveness. Extensive experiments on three benchmark datasets demonstrate consistent improvements over strong baselines while eliminating the need for manual tuning of background and task inference templates.
Semantic identifier learning improves recommendation at scale
FineSID: Scalable and Efficient Semantic Identifier Learning for Generative Recommendation
Abstract: A critical prerequisite of generative recommendation is designing semantic identifiers (SIDs) that are both scalable to large item sets and efficiently learnable. Existing SID learning methods fundamentally rely on Top-1 hard assignment during vector quantization. While heuristic strategies -- such as clustering-based initialization or forced post-hoc collision resolution -- can artificially inflate codebook coverage, they often disrupt end-to-end semantic alignment and fail to address the underlying optimization bottleneck: sparse gradient propagation. In standard Top-1 assignment, gradients concentrate on a narrow subset of frequently selected codewords, leaving the majority inherently under-trained and causing severe SID collisions. To overcome this limitation natively without relying on complex initialization priors, we propose FineSID, a unified quantization framework that moves beyond Top-1 assignment by enabling fine-grained gradient propagation across the entire codebook. Instead of updating only a single selected codeword, FineSID distributes learning signals to all codewords in a soft, differentiable manner. This design promotes globally balanced codebook optimization while strictly preserving semantic consistency, effectively alleviating SID collisions and stabilizing training in large, high-dimensional codebooks. Extensive experiments on multiple public benchmarks demonstrate that FineSID is robust to initialization configurations and consistently improves both codebook utilization and recommendation accuracy. Our work provides a principled, initialization-agnostic solution for semantic identifier learning, advancing the practicality of generative recommendation.
Improving generative recommender accuracy beyond initial predictions
Beyond the Beam: Constructive Repair and Candidate Completion for Generative Recommendation
Abstract: Generative recommenders retrieve items by generating identifiers, but a valid identifier can remain outside the beam after catalog expansion. This raises two connected questions: which failures can identifier assignment repair, and how should retrieval proceed beyond the initial beam? We characterize assignment repair with a fixed generator and retained old identifiers. Output-invariance certificates identify failures shared by all admissible assignments. Under a common effective prefix, coupled support and ranking constraints give the exact feasible interval of new-item counts for target recovery. Building on this characterization, Beyond the Beam (BB) obtains minimum-replacement repairs through an integral flow formulation, selects a shared map and adapts the generator. At inference, generative likelihood and collaborative evidence define one score for ranking, candidate priority and stopping. Retained prefix bounds guide candidate completion and certify its global Top-$K$ when the stopping condition is met. Exhaustive finite-catalog evaluation confirms construction in every feasible case. Across three Amazon Reviews categories and three random seeds, the full T5 procedure improves mean Recall@10 by 15.5--46.3% and NDCG@10 by 15.2--44.4% over the best-performing evaluated generative baseline for each dataset and metric. Matched controls show that shared construction and adaptation improve new-target ranking and certification efficiency on Beauty and Toys. Combined scoring and candidate completion improve NDCG@10 across all three datasets with both T5 and decoder-only LC-Rec.
Audio description generation optimized for timing and content choices
What, When, and How: Audio Description as Constrained Global Optimization
Abstract: Audio Description (AD) makes movies accessible to blind and visually impaired audiences by narrating visual information in gaps between dialogue. Existing automatic AD systems largely treat generation as a local video-to-text problem, assuming that the content to describe and its temporal location are already provided. Realistic AD instead requires coupled decisions about what visual information is narratively important, when it can be spoken without interfering with dialogue, and how it should be formulated to fit within the available time. We formalize AD generation as a constrained optimization problem over these three decisions. Our hybrid system uses large language models to propose and ground visual elements, estimate their salience to the narrative, and generate compressed realizations. A mixed-integer linear program then jointly selects and schedules descriptions across a scene subject to temporal constraints. When evaluated on REFRAMED, a benchmark for realistic AD of movies, our approach makes better decisions than prompted LLMs about what to describe and when to describe it, establishing a new SOTA on narrative QA and temporally grounded metrics. Ablations show that explicit temporal constraints drive gains in placement, while salience estimation controls how much narratively useful content is retained. Improvements are concentrated on temporal and narrative measures rather than n-gram overlap, although a significant gap to professional describers remains.
GNN model adapts recommendations for individuals and groups
A Flexible Recommendation System for Individuals and Groups
Abstract: Group recommender systems typically rely on either aggregating individual preferences or treating groups as distinct meta-users. However, these methods often suffer from static aggregation strategies or data sparsity issues within group histories. This paper introduces a novel approach, that relies on a GNN-based architecture to learn a dual representation of each user's preferences, capturing their behavior as an independent individual from one side and as a member of a collective from the other side. By performing a differential analysis of these individual and group-oriented preferences, our system then determines the behavioral profile of each user when joining a group. Finally, specific preference aggregation strategies are defined to cope with the behavioral profiles of the users composing a group. Consequently, the system is equally capable of delivering precise recommendations to individuals and to arbitrary groups, effectively unifying the two traditional paradigms of recommendation. Experiments on synthetic data simulating diverse group settings and behaviors confirm the flexibility and relevance of the proposed approach compared to state-of-the-art methods.
Generative recommendation improves item suggestions using semantic and collaborative signals
Addressing Cross-Stage Decoupling of Semantic and Collaborative Signals in Generative Recommendation
Abstract: Generative recommendation reformulates sequential recommendation as autoregressive generation by encoding items into semantic tokens, enabling improved scaling capability and cross-domain generalization. However, existing generative recommender systems typically follow a two-stage pipeline, where item tokenization is largely dominated by textual semantics with limited incorporation of collaborative signals and interaction similarity, leading to code assignments that are misaligned with downstream generation. Conversely, the generation stage tends to overlook the original semantic information, as the code sequences are re-embedded based on interaction data. This cross-stage information decoupling limits semantic coherence and recommendation accuracy. To address this issue, we propose SCRec, a general framework that enhances cross-stage coherence through bidirectional information supplementation. Specifically, we introduce (i) collaborative-enhanced tokenization to explicitly inject textualized collaborative signals into semantic tokenization, without introducing additional alignment task, (ii) semantic-guided generation to dynamically recalibrate semantic priors with learnable code embeddings in generation stage, and (iii) manifold alignment to reconcile the geometric mismatch between the embedding space of discrete codebook indices and the dense continuous semantic space. These interrelated components form a general framework that aligns semantic and collaborative signals and enhances cross-stage information coherence, with minimal additional training and inference costs. Extensive experiments demonstrate the effectiveness, robustness, and generalizability of our proposed framework.