15 papers
Teacher Retains Full Tokens, Student Merges Efficiently: TM20K for E-Commerce Sequence Modeling in Ad Recommendation
Xinchun Li, Duoru Zheng, Wenlin Zhao +13
Benefiting from ultra-long behavior sequence modeling, existing recommender systems bring users a better experience via simultaneously considering their long-term and short-term in…
MixFormer: Co-Scaling Up Dense and Sequence in Industrial Recommenders
Xu Huang, Hao Zhang, Zhifang Fan +6
As industrial recommender systems enter a scaling-driven regime, Transformer architectures have become increasingly attractive for scaling models towards larger capacity and longer…
Rec-Distill: An Industrial Distillation Pipeline for Large-Scale Recommendation Models
Haoran Ding, Wenlin Zhao, Yuchen Jiang +16
Large recommendation models have demonstrated substantial potential gains under scaling laws, yet these gains are difficult to realize in industrial recommendation systems because…
Compute Only Once: UG-Separation for Efficient Large Recommendation Models
Hui Lu, Zheng Chai, Shipeng Bai +15
Driven by scaling laws, recommender systems increasingly rely on larger-scale models to capture complex feature interactions and user behaviors, but this trend also leads to prohib…
Bottleneck Tokens for Unified Multimodal Retrieval
Siyu Sun, Jing Ren, Zhaohe Liao +8
Adapting decoder-only multimodal large language models (MLLMs) for unified multimodal retrieval faces two structural gaps. First, existing methods rely on implicit pooling, which o…
TokenMixer-Large: Scaling Up Large Ranking Models in Industrial Recommenders
Yuchen Jiang, Jie Zhu, Xintian Han +18
While scaling laws for recommendation models have gained significant traction, existing architectures such as Wukong, HiFormer and DHEN, often struggle with sub-optimal designs and…