Can Generative Recommendation Reach Cold Items? A Temporal Perspective on Semantic-ID Generation

• URL: http://arxiv.org/abs/2607.21101v1
• Date: Thu, 23 Jul 2026 09:30:49 GMT
• Title: Can Generative Recommendation Reach Cold Items? A Temporal Perspective on Semantic-ID Generation
• Abstract summary: We show that SID-based models can occasionally reach future items supported by observed tokens and prefixes, but struggle with unseen atomic tokens and unsupported SID paths.<n>These findings show that SID generation is compositional but not fully open-ended, and suggest future directions in more independent SID spaces.
• Score: 19.158930665677
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Semantic-ID-based generative recommendation represents items as sequences of shared semantic tokens, enabling token recombination beyond isolated item IDs. However, closed-world recombination does not necessarily imply temporal open-token cold-start induction, where new items enter the item catalog with unseen atomic tokens or weakly supported SID paths. In this work, we revisit SID-based generative recommendation under an absolute-time temporal protocol that separates seen and unseen targets and diagnoses the cold item reachability at the token level. Through seen/unseen-hit analysis, coldness taxonomy, and oracle-prefix probing, we show that current SID-based models can occasionally reach future items supported by observed tokens and prefixes, but struggle with unseen atomic tokens and unsupported SID paths. We further explain this boundary by interpreting SID generation as hierarchical semantic bucketing: early tokens select coarse semantic regions, while later tokens refine item-specific paths. These findings show that SID generation is compositional but not fully open-ended, and suggest future directions in more independent SID spaces, scoring-based interfaces, and dynamic textual context.

Related papers

• EPIC: Explicit Posterior Item Conditioning for Semantic ID Diffusion Recommendation [18.230837696789226]
  Explicit Posterior Item Conditioning introduces explicit item-level competition into SID denoising.<n>EPIC constructs a personalized posterior over feasible candidate items using the current generation context.<n>Experiments on four Amazon benchmarks show consistent improvements over strong baselines.
  arXiv  Detail & Related papers  (2026-09-03T08:21:50Z)
• Rethinking Item Tokenization in Generative Recommenders: From Fixed Atoms to Semantic Subwords [50.621722853036005]
  In generative recommender systems, items are typically tokenized into fixed-length semantic ID sequences for autoregressive next-item prediction.<n>For user-context modeling, this fine-grained representation triggers Intra-item Attention Overload.<n>We propose Semantic Subword Tokenization (SST), which represents historical items as variable-length semantic subwords.
  arXiv  Detail & Related papers  (2026-08-24T02:42:51Z)
• TopoGR: Revealing and Preserving Latent Structure of Semantic ID in Generative Recommendation [25.345503799250476]
  TopoGR is a topology-preserving generative recommendation framework based on Bit-decomposable Semantic ID( Binary SID)<n>We show that TopoGR consistently outperforms existing state-of-the-art baselines in recommendation performance.
  arXiv  Detail & Related papers  (2026-07-28T02:45:33Z)
• CogRec: Structure-Cognitive Fast-and-Slow Reasoning for Generative Recommendation [29.663400191042996]
  CogRec is a structure-cognitive fast-and-slow reasoning framework.<n>CogRec augments the vertical SID hierarchy with intra-layer semantic graphs and item-level neighborhoods.<n> SID Routing represents recommendation reasoning through layer-wise Match, LateralJump, and Explore operations.
  arXiv  Detail & Related papers  (2026-07-27T13:19:55Z)
• DREAM: Dynamic Refinement of Early Assignment Mappings [7.781421530363403]
  Current methods assign each item a single static identifier through offline tokenization before sufficient user feedback is observed.<n>We identify this one-shot commitment as the fundamental cold-start bottleneck in SID-based generative recommendation.<n>We propose DREAM, a three-stage framework that resolves this flaw through progressive refinement.
  arXiv  Detail & Related papers  (2026-06-05T06:21:01Z)
• Conditional Memory Enhanced Item Representation for Generative Recommendation [46.59623438612085]
  Generative recommendation (GR) has emerged as a promising paradigm that predicts target items by autoregressively generating their semantic identifiers (SID)<n>Most GR methods follow a quantization-representation-generation pipeline, first assigning each item a SID, then constructing input representations from SID-token embeddings, and finally predicting the target SID through autoregressive generation.<n>We propose ComeIR, a Conditional Memory enhanced Item Representation framework that reconstructs SID-token embeddings into item-aware inputs and restores the token granularity during SID decoding.
  arXiv  Detail & Related papers  (2026-05-12T02:56:59Z)
• Mitigating Collaborative Semantic ID Staleness in Generative Retrieval [0.0]
  Generative retrieval with Semantic IDs (SIDs) assigns each item a discrete identifier and treats retrieval as a sequence generation problem.<n>As interaction patterns drift over time, SIDs become stale, i.e., their collaborative semantics no longer match recent logs.<n>We propose a lightweight, model-agnostic SID alignment update.
  arXiv  Detail & Related papers  (2026-04-14T20:06:48Z)
• End-to-End Semantic ID Generation for Generative Advertisement Recommendation [33.453121305193434]
  We propose a Unified SID generation framework for generative advertisement recommendation.<n>Specifically, we jointly optimize embeddings and SIDs in an end-to-end manner from raw advertising data.<n>Experiments demonstrate that UniSID consistently outperforms state-of-the-art SID generation methods.
  arXiv  Detail & Related papers  (2026-02-11T02:38:26Z)
• 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)
• Unleashing the Native Recommendation Potential: LLM-Based Generative Recommendation via Structured Term Identifiers [51.64398574262054]
  This paper introduces Term IDs (TIDs), defined as a set of semantically rich and standardized textual keywords, to serve as robust item identifiers.<n>We propose GRLM, a novel framework centered on TIDs, to convert item's metadata into standardized TIDs and utilize Integrative Instruction Fine-tuning to collaboratively optimize term internalization and sequential recommendation.
  arXiv  Detail & Related papers  (2026-01-11T07:53:20Z)
• Order-agnostic Identifier for Large Language Model-based Generative Recommendation [94.37662915542603]
  Items are assigned identifiers for Large Language Models (LLMs) to encode user history and generate the next item.<n>Existing approaches leverage either token-sequence identifiers, representing items as discrete token sequences, or single-token identifiers, using ID or semantic embeddings.<n>We propose SETRec, which leverages semantic tokenizers to obtain order-agnostic multi-dimensional tokens.
  arXiv  Detail & Related papers  (2025-02-15T15:25:38Z)
• Learning Multi-Aspect Item Palette: A Semantic Tokenization Framework for Generative Recommendation [55.99632509895994]
  We introduce LAMIA, a novel approach for multi-aspect semantic tokenization.<n>Unlike RQ-VAE, which uses a single embedding, LAMIA learns an item palette''--a collection of independent and semantically parallel embeddings.<n>Our results demonstrate significant improvements in recommendation accuracy over existing methods.
  arXiv  Detail & Related papers  (2024-09-11T13:49:48Z)
• Language Models As Semantic Indexers [78.83425357657026]
  We introduce LMIndexer, a self-supervised framework to learn semantic IDs with a generative language model. We show the high quality of the learned IDs and demonstrate their effectiveness on three tasks including recommendation, product search, and document retrieval.
  arXiv  Detail & Related papers  (2023-10-11T18:56:15Z)

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.