Preprint
Oct 2026
DuoLM: Fast and Effective Personalized Recommendation with Language Models
DuoLM is introduced, a framework for fast and effective personalized recommendation that integrates a compact recommendation model with an efficient serving runtime that accelerates inference through virtual token IDs for embedding access, shared history computation, and CUDA Graph replay.
Jiang Wu, Yu-Tong Song, Hong-Bo Wang et al.
· 0 citations