查询作为锚点：基于大语言模型的场景自适应用户表征

Query as Anchor: Scenario-Adaptive User Representation via Large Language Model

February 16, 2026

作者: Jiahao Yuan, Yike Xu, Jinyong Wen, Baokun Wang, Ziyi Gao, Xiaotong Lin, Yun Liu, Xing Fu, Yu Cheng, Yongchao Liu, Weiqiang Wang, Zhongle Xie

cs.AI

摘要

工业级用户表征学习需要在稳健的通用性与敏锐的任务敏感性之间取得平衡。然而，现有范式主要生成静态的、任务无关的嵌入表示，难以在统一向量空间中协调下游场景的差异化需求。此外，异构多源数据引入的固有噪声与模态冲突会降低表征质量。我们提出"查询作为锚点"框架，将用户建模从静态编码转向动态的查询感知合成。为赋予大语言模型深度用户理解能力，我们首先构建UserU——一个工业级预训练数据集，将多模态行为序列与用户理解语义对齐；并通过Q-Anchor嵌入架构，将分层粗粒度到细粒度编码器集成至双塔式大语言模型，采用联合对比-自回归优化实现查询感知的用户表征。为弥合通用预训练与专业业务逻辑之间的差距，我们进一步引入基于聚类的软提示调优技术，以强化判别性潜在结构，有效对齐模型注意力与场景特定模态。在部署方面，将查询锚定于序列末端可实现KV缓存加速推理，且增量延迟可忽略不计。在支付宝10个工业基准测试上的评估表明，该方法具有持续的最优性能、强大的可扩展性和高效的部署能力。在支付宝生产系统中针对两个实际场景进行的大规模在线A/B测试进一步验证了其实际有效性。我们的代码已准备公开，即将发布于：https://github.com/JhCircle/Q-Anchor。

English

Industrial-scale user representation learning requires balancing robust universality with acute task-sensitivity. However, existing paradigms primarily yield static, task-agnostic embeddings that struggle to reconcile the divergent requirements of downstream scenarios within unified vector spaces. Furthermore, heterogeneous multi-source data introduces inherent noise and modality conflicts, degrading representation. We propose Query-as-Anchor, a framework shifting user modeling from static encoding to dynamic, query-aware synthesis. To empower Large Language Models (LLMs) with deep user understanding, we first construct UserU, an industrial-scale pre-training dataset that aligns multi-modal behavioral sequences with user understanding semantics, and our Q-Anchor Embedding architecture integrates hierarchical coarse-to-fine encoders into dual-tower LLMs via joint contrastive-autoregressive optimization for query-aware user representation. To bridge the gap between general pre-training and specialized business logic, we further introduce Cluster-based Soft Prompt Tuning to enforce discriminative latent structures, effectively aligning model attention with scenario-specific modalities. For deployment, anchoring queries at sequence termini enables KV-cache-accelerated inference with negligible incremental latency. Evaluations on 10 Alipay industrial benchmarks show consistent SOTA performance, strong scalability, and efficient deployment. Large-scale online A/B testing in Alipay's production system across two real-world scenarios further validates its practical effectiveness. Our code is prepared for public release and will be available at: https://github.com/JhCircle/Q-Anchor.