发表机构
CyberAgent（CyberAgent）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
该研究提出一种按交易模式而非用户进行LLM推理的语义用户画像流水线，显著降低推理成本，并在银行规模部署中实现近三个数量级的效率提升。
AI 中文摘要
对交易历史进行逐用户的大语言模型（LLM）推理，将推理预算与用户数量线性绑定，这在应用规模下变得难以承受。我们将属性推断从逐用户重新构建为逐交易模式。该流水线分三个阶段运行：Resolve阶段通过可选的网络基础（web grounding）抽象化商品名称；Profile阶段为每个频繁模式推断属性；Tag阶段将自由文本属性聚类到可查询的数据库中。在Profile阶段，每个模式仅需一次LLM调用，即可输出预定义的分类标签、自由文本属性以及每个属性的流行度估计。由于推理在模式而非用户上进行，预算随模式数量而非用户数量增长。在公开的Open e-commerce语料库上，该数据库在评估属性上的AUC与直接读取每个用户原始历史的LLM在统计上不可区分，且流行度估计在正负用户之间携带判别性信号。该流水线已部署于一家日本大型银行，对约数千万用户进行画像，与逐用户流水线相比，LLM推理目标减少了近三个数量级。代码已在此https URL上公开。
英文摘要
Per-user LLM inference on transaction histories binds the inference budget linearly to user count, which becomes prohibitive at applied scale. We re-cast attribute inference from per-user to per-transaction-pattern. The pipeline runs in three phases: Resolve abstracts item names with optional web grounding, Profile infers attributes for each frequent pattern, and Tag clusters free-text attributes into a queryable database. In Profile, a single LLM call per pattern emits predefined categorical labels, free-text attributes, and per-attribute prevalence estimates. Because inference runs over patterns rather than users, the budget grows with the pattern count rather than the user count. On the public Open e-commerce corpus, the database is statistically indistinguishable from an LLM that reads each user's raw history directly in AUC across the evaluated attributes, and the prevalence estimates carry discriminative signal between positive and negative users. The pipeline is deployed at a major Japanese bank profiling on the order of tens of millions of users, with close to a three-order-of-magnitude reduction in LLM inference targets versus a per-user pipeline. The code is publicly available on https://github.com/CyberAgentAILab/profiling-agent-open-ecommerce.
Comments10 pages, 3 figures, IEEE International Conference on Data Mining 2026 (ICDM)
