Pctx: Tokenizing Personalized Context for Generative Recommendation
个性化 semantic IDs
2025
RankMixer: Scaling Up Ranking Models in Industrial Recommenders
抖音在大模型推荐上的尝试
Limitations of Dense Retrieval and Beyond
向量检索的瓶颈以及生成式检索的未来
Understanding Embedding Scaling in Collaborative Filtering
理解 Scaling up Embedding Size 过程中会出现的 double-peak/logarithmic 现象
UniSearch: Rethinking Search System with a Unified Generative Architecture
End-to-End 生成式检索
RecBase: Generative Foundation Model Pretraining for Zero-Shot Recommendation
在多个开源数据集上训练的生成式推荐模型, 利用课程学习训练 Tokenizer
Progressive Semantic Residual Quantization for Multimodal-Joint Interest Modeling in Music Recommendation
对残差量化的修改
HLLM-Creator: Hierarchical LLM-based Personalized Creative Generation
利用 LLM 高效生成个性化广告语
R2MR: Review and Rewrite Modality for Recommendation
修正低质量模态特征以改善多模态协同过滤
Representation Quantization for Collaborative Filtering Augmentation
经过正交变换后的向量量化