Gryphon: A Unified Architecture for Semantic-ID Generation and Item-Level Scoring in Industrial Recommendations

• URL: http://arxiv.org/abs/2606.08604v2
• Date: Tue, 09 Jun 2026 13:50:48 GMT
• Title: Gryphon: A Unified Architecture for Semantic-ID Generation and Item-Level Scoring in Industrial Recommendations
• Abstract summary: We introduce a jointly trained item-level scoring component alongside SID generation.<n>Instead of ranking SIDs by accumulated token likelihood, Gryphon resolves each generated SID to its concrete items.<n>On an industrial music service, with item-level scoring trained under a next-item-prediction objective, Gryphon attains the highest item-level Recall@1000.
• Score: 0.0
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Generative retrieval (GR) has become a scalable approach to candidate generation: each item is assigned a short hierarchical token sequence called a Semantic ID (SID), and the next item's SID is decoded autoregressively. A practical limitation is that the decoder's beam search optimizes the likelihood of token sequences, not the relevance of the underlying items. These objectives diverge when sequence likelihood is poorly calibrated due to beam search error accumulation, and when several items collapse onto a single SID and receive identical scores. We introduce Gryphon, an encoder-decoder generative recommendation architecture that adds a jointly trained item-level scoring component alongside SID generation, reusing the encoder's user representation computed in a single forward pass. Instead of ranking SIDs by accumulated token likelihood, Gryphon resolves each generated SID to its concrete items and re-scores those items directly, which sidesteps miscalibrated sequence scores and separates items that collide on the same identifier. On an industrial music service, with item-level scoring trained under a next-item-prediction objective, Gryphon attains the highest item-level Recall@1000, above the strongest baselines (+3.7% over vanilla GR and +2.5% over collision-resolved GR) at comparable parameter count and latency. Gryphon's item-level ranking also surpasses its beam-likelihood ranking of the same candidates (+4.2% gain), demonstrating the benefit of item-level scoring in GR. Deployed as the sole candidate source in a 7-day A/B test, Gryphon produced no statistically significant change in total listening time (+0.25%) while replacing a pipeline of more than 15 candidate generators and a separate preranking stage, substantially simplifying the candidate-generation system.

Related papers

• CodeHID: Learning an Addressable Hierarchical Code Index for Generative Code Retrieval [16.418964135916433]
  We propose CodeHID, a generative code retrieval framework that reformulates the code retrieval task from flat candidate matching to coarse-to-fine semantic address generation.<n>CodeHID relies on two core components. First, Pseudo-Neighbor Guided DocID Learning constructs a globally static hierarchical index by applying multi-level residual quantization and $k$-nearest-neighbor pseudo-supervision.<n>Second, Dual-Phase DocID Generation Guidance reliably navigates this fixed index by combining training-side ranking enhancements, using hard negatives and rank distillation, with inference-side candidate constraints and prefix-aware decoding.
  arXiv  Detail & Related papers  (2026-08-25T05:35:17Z)
• Hierarchical Residual Policy Optimization for Generative Recommendations [46.030303420508254]
  We propose a post-training framework that converts item-level outcomes into dense, token-aligned learning signals.<n>Experiments on a public dataset and an online A/B test in a large-scale commercial system show consistent gains in session-level utility and key business metrics.
  arXiv  Detail & Related papers  (2026-08-01T16:33:15Z)
• Closing the Indexing-Decoding Gap in Multimodal Generative Retrieval via Prefix Retention Optimization [68.48718919047127]
  Multimodal generative retrieval formulates multimodal retrieval as discrete identifier generation, eliminating the need for explicit similarity search over external embeddings.<n>Existing approaches construct identifiers via residual quantization and decode them with trie-constrained beam search.<n>This combination introduces an indexing-decoding gap: identifier learning objectives, including reconstruction and contrastive losses, do not explicitly enforce prefix discriminability during decoding.<n>We propose PRO, a unified framework comprising three mechanisms: (i) prefix ranking distillation aligns quantized prefixes with those induced by pre-quantization embeddings using a listwise loss; (ii) vocabulary scheduling increases codebook
  arXiv  Detail & Related papers  (2026-06-08T09:15:47Z)
• How Reliable Are Semantic-ID Tokenizer Comparisons in Generative Recommendation? [3.9133310867939177]
  In generative recommendation, each item is represented as a sequence of discrete codes, and an autoregressive model is trained to generate the SID sequence of the next item.<n>This evaluation protocol equates SID-level matching with item-level recommendation, an equivalence that holds only when every SID sequence maps to a single item.<n>We show this assumption breaks down in practice: because tokenizers compress item features into a code space, semantically similar but collaboratively distinct items are frequently assigned the same SID sequence.
  arXiv  Detail & Related papers  (2026-05-25T01:22:45Z)
• F-GRPO: Factorized Group-Relative Policy Optimization for Unified Candidate Generation and Ranking [79.49893545611779]
  Large Language Models (LLMs) can generate a subset and order it within a single autoregressive pass.<n>This flexibility introduces a new optimization challenge: the model must search an output space while receiving utility feedback only after the full ranked list is generated.<n>This credit assignment gap makes end-to-end optimization unstable and sample-inefficient.<n>We propose a unified framework that performs both within a single autoregressive rollout and optimize them end-to-end via factorized group-relative policy optimization.
  arXiv  Detail & Related papers  (2026-05-13T04:52:33Z)
• CapsID: Soft-Routed Variable-Length Semantic IDs for Generative Recommendation [6.439016735163573]
  Generative recommendation maps each item to a sequence of Semantic IDs (SIDs) and recasts retrieval as autoregressive token generation.<n> CAPSID replaces hard residual quantization with capsule routing.<n>On Amazon Beauty, Sports, Toys, and a 35M-item proprietary industrial catalog, CAPSID+SEMANTICBPE improves Recall at 10 by 9.6% on average over ReSID.
  arXiv  Detail & Related papers  (2026-05-06T16:33:13Z)
• From Local Indices to Global Identifiers: Generative Reranking for Recommender Systems via Global Action Space [50.72071213515985]
  GloRank is a generative framework that shifts reranking from selecting local indices to generating global identifiers.<n>We show that GloRank consistently outperforms state-of-the-art baselines and achieves superior robustness in cold-start scenarios.
  arXiv  Detail & Related papers  (2026-04-28T06:57:00Z)
• GenRec: A Preference-Oriented Generative Framework for Large-Scale Recommendation [14.663434490160016]
  We present GenRec, a preference-oriented generative framework deployed on the JD App.<n>In month-long online A/B tests, GenRec achieves 9.5% improvement in click count and 8.7% in transaction count over the existing pipeline.
  arXiv  Detail & Related papers  (2026-04-16T11:07:05Z)
• UniRec: Bridging the Expressive Gap between Generative and Discriminative Recommendation via Chain-of-Attribute [12.89120699793625]
  Generative Recommendation (GR) reframes retrieval and ranking as autoregressive decoding over Semantic IDs (SIDs)<n>We formalize this via Bayes' theorem: ranking by p(y|f,u) is equivalent to ranking by p(f|y,u), which factorizes autoregressively over item features.<n>We propose UniRec with Chain-of-Attribute (CoA) as its core mechanism. CoA prefixes each SID sequence with structured attribute tokens:category, seller, brand, before decoding the SID, recovering the item-side feature crossing that discriminative models exploit.
  arXiv  Detail & Related papers  (2026-04-14T03:13:50Z)
• Differentiable Semantic ID for Generative Recommendation [65.83703273297492]
  Generative recommendation provides a novel paradigm in which each item is represented by a discrete semantic ID (SID) learned from rich content.<n>In practice, SIDs are typically optimized only for content reconstruction rather than recommendation accuracy.<n>A natural approach is to make semantic indexing differentiable so that recommendation gradients can directly influence SID learning.<n>We propose DIGER, a first step toward effective differentiable semantic IDs for generative recommendation.
  arXiv  Detail & Related papers  (2026-01-27T15:34:11Z)
• DiffGRM: Diffusion-based Generative Recommendation Model [63.35379395455103]
  Generative recommendation (GR) is an emerging paradigm that represents each item via a tokenizer as an n-digit semantic ID (SID)<n>We propose DiffGRM, a diffusion-based GR model that replaces the autoregressive decoder with a masked discrete diffusion model (MDM)<n> Experiments show consistent gains over strong generative and discriminative recommendation baselines on multiple datasets.
  arXiv  Detail & Related papers  (2025-10-21T03:23:32Z)
• Recommender Systems with Generative Retrieval [58.454606442670034]
  We propose a novel generative retrieval approach, where the retrieval model autoregressively decodes the identifiers of the target candidates. To that end, we create semantically meaningful of codewords to serve as a Semantic ID for each item. We show that recommender systems trained with the proposed paradigm significantly outperform the current SOTA models on various datasets.
  arXiv  Detail & Related papers  (2023-05-08T21:48:17Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.