{"total":12,"items":[{"citing_arxiv_id":"2606.31031","ref_index":1,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"GenPage: Towards End-to-End Generative Homepage Construction at Netflix","primary_cat":"cs.IR","submitted_at":"2026-06-30T02:00:05+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"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.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2606.00422","ref_index":1,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"UniPinRec: Unifying Generative Retrieval and Ranking at Pinterest Scale","primary_cat":"cs.IR","submitted_at":"2026-05-29T23:17:34+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"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.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2606.00324","ref_index":1,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"LLMs Need Encoders for Semantic IDs Too","primary_cat":"cs.IR","submitted_at":"2026-05-29T20:01:06+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":7.0,"formal_verification":"none","one_line_summary":"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.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.14512","ref_index":1,"ref_count":1,"confidence":0.9,"is_internal_anchor":true,"paper_title":"Asymmetric Generative Recommendation via Multi-Expert Projection and Multi-Faceted Hierarchical Quantization","primary_cat":"cs.IR","submitted_at":"2026-05-14T07:55:43+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":7.0,"formal_verification":"none","one_line_summary":"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.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2604.24472","ref_index":2,"ref_count":1,"confidence":0.9,"is_internal_anchor":true,"paper_title":"Modeling Behavioral Intensity and Transitions for Generative Recommendation","primary_cat":"cs.IR","submitted_at":"2026-04-27T13:40:35+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"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.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2604.13468","ref_index":1,"ref_count":1,"confidence":0.9,"is_internal_anchor":true,"paper_title":"From Relevance to Authority: Authority-aware Generative Retrieval in Web Search Engines","primary_cat":"cs.IR","submitted_at":"2026-04-15T04:44:45+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"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.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2604.13273","ref_index":1,"ref_count":1,"confidence":0.9,"is_internal_anchor":true,"paper_title":"Mitigating Collaborative Semantic ID Staleness in Generative Retrieval","primary_cat":"cs.IR","submitted_at":"2026-04-14T20:06:48+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"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.","context_count":1,"top_context_role":"background","top_context_polarity":"unclear","context_text":"This research is financially supported by the Foundation for Na- tional Technology Initiative's Projects Support as a part of the roadmap implementation for the development of the high-tech field of Artificial Intelligence for the period up to 2030 (agreement 70-2021-00187) SIGIR '26, July 20-24, 2026, Melbourne, VIC, Australia. Baikalov et al. References [1] Prabhat Agarwal, Anirudhan Badrinath, Laksh Bhasin, Jaewon Yang, Edoardo Botta, Jiajing Xu, and Charles Rosenberg. 2025. PinRec: Outcome-Conditioned, Multi-Token Generative Retrieval for Industry-Scale Recommendation Systems. arXiv:2504.10507 [cs.IR] https://arxiv.org/abs/2504.10507 [2] Fedor Borisyuk, Qingquan Song, Mingzhou Zhou, Ganesh Parameswaran, Madhu"},{"citing_arxiv_id":"2604.07739","ref_index":1,"ref_count":1,"confidence":0.9,"is_internal_anchor":true,"paper_title":"Efficient Dataset Selection for Continual Adaptation of Generative Recommenders","primary_cat":"cs.IR","submitted_at":"2026-04-09T02:48:52+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":4.0,"formal_verification":"none","one_line_summary":"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.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2604.04976","ref_index":1,"ref_count":1,"confidence":0.9,"is_internal_anchor":true,"paper_title":"Tencent Advertising Algorithm Challenge 2025: All-Modality Generative Recommendation","primary_cat":"cs.IR","submitted_at":"2026-04-04T17:05:15+00:00","verdict":"ACCEPT","verdict_confidence":"MODERATE","novelty_score":7.0,"formal_verification":"none","one_line_summary":"Releases TencentGR-1M and TencentGR-10M datasets with baselines for all-modality generative recommendation in advertising, including weighted evaluation for conversions.","context_count":1,"top_context_role":"background","top_context_polarity":"background","context_text":"based user interest modeling [3, 4, 8, 9, 15, 19, 21, 45, 70-72]. Build- ing on these advances, recommender systems in large-scale plat- forms are now increasingly moving from discriminative formu- lations to generative architectures that operate directly on user behavior sequences [12, 18, 29, 53, 56, 60, 67-69]. Instead of merely re-scoring a fixed candidate set, recent genera- tive recommendation models [1, 23, 46, 61] reformulate retrieval or ranking as sequence generation over item identifiers or semantic codes. These models focus on the following intertwined design axes: (i) how to organize the data which consists of both non-sequential tokens such as users' demographic features, as well as heteroge- neous sequential tokens, including both the interacted item token"},{"citing_arxiv_id":"2604.02684","ref_index":1,"ref_count":1,"confidence":0.9,"is_internal_anchor":true,"paper_title":"MBGR: Multi-Business Prediction for Generative Recommendation at Meituan","primary_cat":"cs.IR","submitted_at":"2026-04-03T03:26:36+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"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.","context_count":1,"top_context_role":"background","top_context_polarity":"background","context_text":"RPG [7] proposed a parallel semantic ID generation framework, using product quantization to support the unordered parallel gen- eration of long semantic IDs. The latest work has begun to focus on commercial applications. EGA-V2 [24] extends the generative architecture to the advertising system, uniformly handling multiple links such as bidding and creative selection; PinRec [1] proposes a result-oriented generation method, meeting the needs of different business goals through conditional generation. 2.2 Multi-business recommendations Multi-business (scenario) recommendation systems aim to lever- age data from multiple scenarios to train a unified model to im- prove recommendation effectiveness. Early research, starting from"},{"citing_arxiv_id":"2510.21242","ref_index":1,"ref_count":1,"confidence":0.9,"is_internal_anchor":true,"paper_title":"Bi-Level Optimization for Generative Recommendation: Bridging Tokenization and Generation","primary_cat":"cs.IR","submitted_at":"2025-10-24T08:25:56+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"BLOGER is a bi-level optimization framework that jointly optimizes the tokenizer and recommender for generative recommendation, outperforming prior methods on real-world datasets.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2508.20900","ref_index":3,"ref_count":1,"confidence":0.9,"is_internal_anchor":true,"paper_title":"OneRec-V2 Technical Report","primary_cat":"cs.IR","submitted_at":"2025-08-28T15:29:51+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"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.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null}],"limit":50,"offset":0}