PrefixMem encoder for Semantic IDs improves deepest-level accuracy by up to 46% relative and full-SID retrieval recall by up to 22% relative on Pinterest data across LLM families.
hub
PinRec: Unified Generative Retrieval for Pinterest Recommender Systems
12 Pith papers cite this work. Polarity classification is still indexing.
12
Pith papers citing it
abstract
Generative retrieval methods employ sequential modeling techniques, like transformers, to generate candidate items for recommender systems. These methods have demonstrated promising results in academic benchmarks, surpassing traditional retrieval models such as two-tower architectures. However, a key limitation is that current approaches require a separate model for each product surface, as building a unified model that accommodates the different business needs of various surfaces has proven challenging. Furthermore, existing methods often fail to capture the evolution of user interests over a sequence, focusing instead on only predicting the next item. This paper introduces Pinrec, a novel unified generative retrieval model for all of Pinterest's recommendation surfaces, including home feed, search, and related pins. Pinrec is pretrained on user activity sequences aggregated across surfaces, then fine-tuned for each surface using that surface's impression data. This pretraining-fine-tuning approach enables a single unified model while still adapting to the needs of individual surfaces. To better align recommendations with surface-specific business goals, Pinrec incorporates a novel outcome-conditioned generation mechanism that targets different outcomes for each surface, which further enhances the impact of fine-tuning. Our experiments show that Pinrec balances performance, diversity, and efficiency, delivering significant gains such as +4% increase in search saves. To our knowledge, this paper presents the first rigorous study of a unified generative retrieval model built and deployed at Pinterest scale, marking a significant milestone in the field.
hub tools
citation-role summary
background 3
citation-polarity summary
fields
cs.IR 12roles
background 3representative citing papers
AsymRec decouples input and output representations in generative recommendation via multi-expert semantic projection and multi-faceted hierarchical quantization, outperforming prior models by 15.8% on average.
Releases TencentGR-1M and TencentGR-10M datasets with baselines for all-modality generative recommendation in advertising, including weighted evaluation for conversions.
GenPage is a transformer that autoregressively generates entire structured Netflix homepages from user prompts, delivering +0.24% engagement lift and 20% latency reduction versus production baseline in online tests.
UniPinRec unifies retrieval and ranking into a single model and pipeline deployed at Pinterest, reporting +1% engagement lift, 11.1% lower latency, and 63.6% higher QPS.
BITRec improves generative multi-behavior recommendation by modeling behavioral intensity via separated pathways and transitions via learnable relation matrices, reporting 15-23% gains on large retail datasets.
AuthGR is the first generative retriever to explicitly incorporate document authority alongside relevance using multimodal scoring and progressive training, yielding efficiency gains and real-world engagement improvements.
MBGR is a new generative recommendation framework using business-aware semantic IDs, multi-business prediction, and label dynamic routing to handle multiple businesses without seesaw effects or representation confusion, validated by experiments and deployed at Meituan.
BLOGER is a bi-level optimization framework that jointly optimizes the tokenizer and recommender for generative recommendation, outperforming prior methods on real-world datasets.
A model-agnostic SID alignment update mitigates staleness from temporal drift in user-item interactions for generative retrievers, improving Recall@K and nDCG@K while reducing compute by 8-9x versus full retraining.
OneRec-V2 scales generative recommendation to 8B parameters via decoder-only design and real-world preference alignment, improving user engagement metrics in production A/B tests.
Gradient-based representations paired with distribution-matching enable efficient curation of small data subsets that improve performance and training efficiency for continually adapting generative recommenders while maintaining robustness to distributional drift.
citing papers explorer
-
LLMs Need Encoders for Semantic IDs Too
PrefixMem encoder for Semantic IDs improves deepest-level accuracy by up to 46% relative and full-SID retrieval recall by up to 22% relative on Pinterest data across LLM families.
-
Asymmetric Generative Recommendation via Multi-Expert Projection and Multi-Faceted Hierarchical Quantization
AsymRec decouples input and output representations in generative recommendation via multi-expert semantic projection and multi-faceted hierarchical quantization, outperforming prior models by 15.8% on average.
-
GenPage: Towards End-to-End Generative Homepage Construction at Netflix
GenPage is a transformer that autoregressively generates entire structured Netflix homepages from user prompts, delivering +0.24% engagement lift and 20% latency reduction versus production baseline in online tests.
-
UniPinRec: Unifying Generative Retrieval and Ranking at Pinterest Scale
UniPinRec unifies retrieval and ranking into a single model and pipeline deployed at Pinterest, reporting +1% engagement lift, 11.1% lower latency, and 63.6% higher QPS.
-
Modeling Behavioral Intensity and Transitions for Generative Recommendation
BITRec improves generative multi-behavior recommendation by modeling behavioral intensity via separated pathways and transitions via learnable relation matrices, reporting 15-23% gains on large retail datasets.
-
From Relevance to Authority: Authority-aware Generative Retrieval in Web Search Engines
AuthGR is the first generative retriever to explicitly incorporate document authority alongside relevance using multimodal scoring and progressive training, yielding efficiency gains and real-world engagement improvements.
-
MBGR: Multi-Business Prediction for Generative Recommendation at Meituan
MBGR is a new generative recommendation framework using business-aware semantic IDs, multi-business prediction, and label dynamic routing to handle multiple businesses without seesaw effects or representation confusion, validated by experiments and deployed at Meituan.
-
Bi-Level Optimization for Generative Recommendation: Bridging Tokenization and Generation
BLOGER is a bi-level optimization framework that jointly optimizes the tokenizer and recommender for generative recommendation, outperforming prior methods on real-world datasets.
-
Mitigating Collaborative Semantic ID Staleness in Generative Retrieval
A model-agnostic SID alignment update mitigates staleness from temporal drift in user-item interactions for generative retrievers, improving Recall@K and nDCG@K while reducing compute by 8-9x versus full retraining.
-
OneRec-V2 Technical Report
OneRec-V2 scales generative recommendation to 8B parameters via decoder-only design and real-world preference alignment, improving user engagement metrics in production A/B tests.
-
Efficient Dataset Selection for Continual Adaptation of Generative Recommenders
Gradient-based representations paired with distribution-matching enable efficient curation of small data subsets that improve performance and training efficiency for continually adapting generative recommenders while maintaining robustness to distributional drift.