Generative Recommendation
PulseAugur coverage of Generative Recommendation — every cluster mentioning Generative Recommendation across labs, papers, and developer communities, ranked by signal.
2 天有情绪数据
-
新的UniSID框架通过端到端SID生成改进广告推荐
研究人员开发了UniSID，一个用于在生成式推荐系统中生成语义ID（SIDs）的新颖框架，特别适用于广告推荐。这种新方法通过端到端联合优化嵌入和SID，解决了现有方法（如Residual Quantization）的局限性，防止了语义退化和误差累积。UniSID还采用了一种多粒度对比学习策略和一种基于摘要的广告重建机制来捕捉更细粒度的语义。实验表明，在下游广告场景中，与当前最先进的方法相比，UniSID的命中率指标提高了4.62%。
-
New framework ComeIR boosts generative recommendation systems
Researchers have introduced ComeIR, a new framework designed to enhance generative recommendation systems. This approach addresses challenges in how item representations are constructed, aiming to preserve crucial struc…
-
Researchers propose new framework for generative recommendation systems
Researchers have developed a new framework to improve the generation of Semantic IDs (SIDs) for generative recommendation systems. This approach addresses issues of information and semantic degradation by integrating de…