Semantics Meet Signals: Dual Codebook Representationl Learning for Generative Recommendation

TL;DR

FlexCode introduces a dual codebook approach for generative recommendation, balancing collaborative signals and semantic understanding to improve accuracy and robustness across item popularity levels.

Contribution

The paper proposes FlexCode, a popularity-aware dual codebook framework with a dynamic MoE, enhancing representational efficiency and generalization in generative recommendation models.

Findings

Outperforms strong baselines on public and industrial datasets.

Improves accuracy and tail robustness in recommendations.

Balances memorization and generalization effectively.

Abstract

Generative recommendation has recently emerged as a powerful paradigm that unifies retrieval and generation, representing items as discrete semantic tokens and enabling flexible sequence modeling with autoregressive models. Despite its success, existing approaches rely on a single, uniform codebook to encode all items, overlooking the inherent imbalance between popular items rich in collaborative signals and long-tail items that depend on semantic understanding. We argue that this uniform treatment limits representational efficiency and hinders generalization. To address this, we introduce FlexCode, a popularity-aware framework that adaptively allocates a fixed token budget between a collaborative filtering (CF) codebook and a semantic codebook. A lightweight MoE dynamically balances CF-specific precision and semantic generalization, while an alignment and smoothness objective maintains coherence across the popularity spectrum. We perform experiments on both public and industrial-scale datasets, showing that FlexCode consistently outperform strong baselines. FlexCode provides a new mechanism for token representation in generative recommenders, achieving stronger accuracy and tail robustness, and offering a new perspective on balancing memorization and generalization in token-based recommendation models.