CRAB: Codebook Rebalancing for Bias Mitigation in Generative Recommendation

2026-04-06

AI Summary

This work addresses the issue of popularity bias in generative recommender systems, where the skewed distribution of semantic tokens disproportionately favors popular items and degrades the representation quality of long-tail items. To mitigate this without retraining the underlying model, the authors propose a post-processing debiasing approach that rebalances token frequencies through codebook adjustment and preserves hierarchical semantic consistency via tree-structured regularization over discrete token representations. Evaluated on real-world datasets, the method effectively alleviates popularity bias while simultaneously improving overall recommendation accuracy, achieving a favorable trade-off between fairness and performance.

Abstract

Generative recommendation (GeneRec) has introduced a new paradigm that represents items as discrete semantic tokens and predicts items in a generative manner. Despite its strong performance across multiple recommendation tasks, existing GeneRec approaches still suffer from severe popularity bias and may even exacerbate it. In this work, we conduct a comprehensive empirical analysis to uncover the root causes of this phenomenon, yielding two core insights: 1) imbalanced tokenization inherits and can further amplify popularity bias from historical item interactions; 2) current training procedures disproportionately favor popular tokens while neglecting semantic relationships among tokens, thereby intensifying popularity bias. Building on these insights, we propose CRAB, a post-hoc debiasing strategy for GeneRec that alleviates popularity bias by mitigating frequency imbalance among semantic tokens. Specifically, given a well-trained model, we first rebalance the codebook by splitting over-popular tokens while preserving their hierarchical semantic structure. Based on the adjusted codebook, we further introduce a tree-structured regularizer to enhance semantic consistency, encouraging more informative representations for unpopular tokens during training. Experiments on real-world datasets demonstrate that CRAB significantly improves recommendation performance by effectively alleviating popularity bias.

Problem

• popularity bias
• generative recommendation
• codebook imbalance
• semantic tokens
• recommendation fairness

Innovation

• Generative Recommendation
• Popularity Bias
• Codebook Rebalancing
• Semantic Tokenization
• Tree-structured Regularization