SIGIR

FLAME: Condensing Ensemble Diversity into a Single Network for Efficient Sequential Recommendation

FLAME：将集成多样性压缩到单一网络以实现高效的序列推荐

WooJoo Kim, JunYoung Kim, JaeHyung Lim, SeongJin Choi, SeongKu Kang, HwanJo Yu

AI总结 FLAME通过将集成多样性压缩到单一网络，提升序列推荐效率。利用模块化集成和预训练冻结网络，实现稳定优化，实验显示在六个数据集上优于现有方法。

Comments Accepted to SIGIR 2026 full papers track