用两个网络模拟海量多样性，让单模型实现高效推荐
FLAME: Condensing Ensemble Diversity into a Single Network for Efficient Sequential Recommendation
- 仅用两个网络通过模块动态组合，生成丰富表征模式
- 训练快7.69倍，NDCG@20提升9.70%，推理零开销
- 适合追求高效率与强表现的工业级推荐系统
序列推荐需捕捉用户行为多样性，单一网络常力不从心。尽管集成方法能缓解此问题，但从头训练多个网络带来高计算成本及噪声互监督导致的不稳定性。本文提出FLAME框架——冻结与可学习模块化集成，将集成级别的多样性浓缩至单个网络中实现高效序列推荐。训练阶段，FLAME仅用两个网络通过模块化集成模拟指数级多样性，动态组合各网络子模块（如层）以生成丰富的表征模式；为稳定训练，先预训练并冻结一个网络作为语义锚点，并采用引导式互学习将其余可学习网络的多样化表征对齐至该锚点空间；推理时仅使用可学习网络，实现集成级性能且相比单网络无额外开销。六组数据集实验表明，FLAME优于现有最先进基线，收敛速度最快达7.69倍提升，NDCG@20最高提高9.70%。代码已开源：https://github.com/woo-joo/FLAME_SIGIR26。
原文摘要 · Abstract (English)
Sequential recommendation requires capturing diverse user behaviors, which a single network often fails to capture. While ensemble methods mitigate this, training multiple networks from scratch incurs high computational cost and instability from noisy mutual supervision. We propose Frozen and Learnable networks with Aligned Modular Ensemble (FLAME), a novel framework that condenses ensemble-level diversity into a single network for efficient sequential recommendation. During training, FLAME simulates exponential diversity using only two networks via modular ensemble, which dynamically combines sub-modules (e.g., layers) of each network to generate a rich space of diverse representation patterns. To stabilize training, FLAME pretrains and freezes one network as a semantic anchor and employs guided mutual learning to align diverse representations into the space of remaining learnable network. At inference, FLAME utilizes only the learnable network, achieving ensemble-level performance with zero overhead compared to a single network. Experiments on six datasets show that FLAME outperforms state-of-the-art baselines, achieving up to 7.69x faster convergence and 9.70% improvement in NDCG@20. Our code is available at https://github.com/woo-joo/FLAME_SIGIR26.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。