Scaling large recommendation systems requires advancing three major frontiers: processing longer user histories, expanding candidate sets, and increasing model capacity. While promising, transformers' computational cost scales quadratically with the user sequence length and linearly with the number of candidates. This trade-off makes it prohibitively expensive to expand candidate sets or increase sequence length at inference, despite the significant performance improvements. We introduce \textbf{LIME}, a novel architecture that resolves this trade-off. Through two key innovations, LIME fundamentally reduces computational complexity. First, low-rank ``link embeddings" enable pre-computation of attention weights by decoupling user and candidate interactions, making the inference cost nearly independent of candidate set size. Second, a linear attention mechanism, \textbf{LIME-XOR}, reduces the complexity with respect to user sequence length from quadratic ($O(N^2)$) to linear ($O(N)$). Experiments on public and industrial datasets show LIME achieves near-parity with state-of-the-art transformers but with a 10$\times$ inference speedup on large candidate sets or long sequence lengths. When tested on a major recommendation platform, LIME improved user engagement while maintaining minimal inference costs with respect to candidate set size and user history length, establishing a new paradigm for efficient and expressive recommendation systems.

翻译：扩展大规模推荐系统需要攻克三大前沿挑战：处理更长的用户历史记录、扩展候选集规模以及提升模型容量。尽管Transformer模型展现出巨大潜力，但其计算成本随用户序列长度呈二次方增长，并随候选物品数量线性增加。这种权衡关系使得在推理阶段扩展候选集或增加序列长度变得极其昂贵，尽管这些操作能带来显著的性能提升。本文提出\textbf{LIME}这一新颖架构，从根本上解决了这一权衡问题。通过两项关键创新，LIME显著降低了计算复杂度。首先，低秩“链接嵌入”通过解耦用户与候选物品的交互，实现了注意力权重的预计算，使推理成本几乎与候选集规模无关。其次，线性注意力机制\textbf{LIME-XOR}将用户序列长度的计算复杂度从二次方（$O(N^2)$）降低至线性（$O(N)$）。在公开数据集和工业数据集上的实验表明，LIME在达到与最先进Transformer模型近乎同等性能的同时，能在大型候选集或长序列场景下实现10倍的推理加速。在主流推荐平台的实际测试中，LIME在保持候选集规模和用户历史长度相关推理成本最小化的同时，有效提升了用户参与度，为高效且高表达的推荐系统建立了新范式。