Alimama Tech
May 27, 2021 · Artificial Intelligence
Advances in Click‑Through Rate (CTR) Modeling: Overview of Recent SIGIR Papers and Optimization Paths
The article reviews recent Alibaba Mama advances in click‑through‑rate modeling, classifying optimizations across the three‑layer CTR architecture and highlighting three SIGIR papers—GIN’s graph‑based user intent modeling, PCF’s pre‑trained GNN for explicit cross‑feature semantics, and FSCD’s compute‑factor‑guided automatic feature selection—each boosting prediction accuracy and system efficiency.
CTR predictionGraph Neural NetworksMachine Learning
0 likes · 12 min read