ComeIR：条件记忆增强的生成式推荐物品表示方法
Conditional Memory Enhanced Item Representation for Generative Recommendation
AI 导读
研究提出 ComeIR，一种条件记忆增强的物品表示框架，用于生成式推荐（GR）中连接量化与生成的表示阶段，解决物品语义丢失和结构化 SID 关系建模不足两大问题。
正文 · 原文
Abstract:Generative recommendation (GR) has emerged as a promising paradigm that predicts target items by autoregressively generating their semantic identifiers (SID). Most GR methods follow a quantization-representation-generation pipeline. While most research focuses on quantization and generation, the representation that bridges the two remains underexplored. Existing representation constructions mainly take two forms: either directly flattening the SID code embeddings into a sequence or merging them into a compact vector via external networks. While effective, they still expose two practical challenges: \textbf{1) Loss of item semantics:} Information lost by quantization or compression needs to be recovered during the representation stage. \textbf{2) Insufficient modeling of structured SID relations:} Intra-item code composition patterns and inter-item prefix-transition patterns are not explicitly preserved for representation and subsequent SID generation.
To overcome these challenges, we propose \textbf{ComeIR}, a \textbf{\uline{Co}}nditional \textbf{\uline{me}}mory Enhanced \textbf{\uline{I}}tem \textbf{\uline{R}}epresentation framework that reconstructs SID code embeddings into item-aware inputs and preserves structured SID evidence for prediction. Item-guided Token Scoring estimates the contribution of each SID code using the item embedding employed by the quantizer. Dual-level Conditional Memory captures intra-item prefix composition and inter-item prefix transitions, while Memory Merge integrates these signals into each historical item representation. The Memory-restoring Prediction Head further conditions SID-level prediction on the generated prefix and historical transition evidence. Extensive experiments demonstrate the effectiveness and flexibility of ComeIR, and further reveal scalable gains from enlarging conditional memory.
| Subjects: | Information Retrieval (cs.IR); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2605.11447 [cs.IR] |
| (or arXiv:2605.11447v2 [cs.IR] for this version) | |
| https://doi.org/10.48550/arXiv.2605.11447 arXiv-issued DOI via DataCite |
Submission history
From: Ziwei Liu [view email]
[v1]
Tue, 12 May 2026 02:56:59 UTC (538 KB)
[v2]
Wed, 28 Oct 2026 09:28:27 UTC (1,037 KB)
来源：arXiv cs.AI · arxiv.org