CRAB: Codebook Rebalancing for Bias Mitigation in Generative Recommendation

1. 1Conduct empirical analysis to identify root causes of popularity bias in generative recommendation systems
2. 2Develop codebook rebalancing technique to mitigate bias
3. 3Implement CRAB algorithm to rebalance codebooks and improve recommendation diversity

Machine learning engineers and researchers working on recommendation systems can benefit from this approach to improve the fairness and diversity of their models

💡 Popularity bias in generative recommendation systems can be mitigated by rebalancing codebooks

💡 CRAB: Codebook Rebalancing for Bias Mitigation in Generative Recommendation