This article reviews Huawei Noah's Ark Lab's exploration of large language models in recommendation systems, covering background challenges, the KAR and Uni-CTR projects, experimental results, and future research directions for open, knowledge‑driven recommendation pipelines.
The paper presents a comprehensive multi‑scenario recommendation study introducing three models—SACN, SAINet, and DSWIN—that integrate scene‑aware attention, attribute‑level preferences, and contrastive disentanglement to capture distinct user interests, achieving consistent AUC gains and online CTR improvements across real‑world datasets.
AdaScene introduces an adaptive scenario modeling framework for Meituan’s DSP that mitigates negative transfer and extreme data sparsity across heterogeneous display scenes by employing a knowledge‑transfer network with scene‑specific feature adaptation and gated expert sharing, alongside a gradient‑based scene‑aggregation process that clusters similar scenarios, yielding consistent performance gains for both high‑traffic and low‑traffic channels.
