提出QGS模型，让搜索推荐更懂用户当前查询意图。
From Item-Only to Query-Item: Query-Conditioned Generative Search with QGS in Quark
- 用（查询,物品）对建模交互，区分查询边界
- 线上测试提升点击率0.62%，停留时长增加3.55%
- 适合做高精度搜索推荐的工程师和研究者
生成式序列模型在推荐中表现优异，但在搜索排序中面临挑战。搜索行为本质由查询驱动，每次查询切换都会带来用户交互历史的剧烈主题变化。现有生成方法将查询与物品混合为单一序列，不区分查询边界，导致模型混淆不同查询意图，产生噪声监督。本文提出查询条件生成搜索（QGS），将每个交互表示为（查询, 物品）对令牌，并采用查询条件下的下一物品预测目标。预测目标从模糊的边缘分布 P(item_{t+1}|context_{<=t}) 变为清晰的条件分布 P(item_{t+1}|context_{<=t}, query_{t+1})，直接消除查询切换带来的语义断层。标准注意力编码长序列复杂度为二次方，难以满足严格在线延迟要求。本文引入线性HSTU编码器，以因果线性递推替代全注意力，每层复杂度从 O(L^2) 降至 O(L)，且排名质量无损。传统搜索依赖人工特征如文本匹配分数、统计信号和行为特征。本文提出HFG-Attention，将异构特征按语义分组并通过专用注意力模块融合，实现稀疏工程信号与密集序列表示的衔接。QGS已部署于中国主流搜索引擎Quark Search的排序模块。在线A/B测试显示显著提升：点击率 +0.62%，点击-搜索比 +0.38%，页面浏览时长 +3.55%，优于现有深度学习基线。
原文摘要 · Abstract (English)
Generative sequence models have shown strong results in recommendation. Applying them to search ranking is more challenging. Search behavior is inherently query-driven. Each query switch introduces a sharp topic shift in the user's interaction history. Existing generative methods flatten queries and items into a single token sequence. They do not distinguish query boundaries. This causes the model to mix different query intents into one prediction target, resulting in noisy supervision. We present Query-Conditioned Generative Search (QGS). QGS encodes each interaction as a (query, item) pair token. It trains with a query-conditioned next-item objective. The prediction target changes from a noisy marginal P(item_{t+1}|context_{<=t}) to a clean conditional P(item_{t+1}|context_{<=t}, query_{t+1}). This directly removes the semantic discontinuity caused by query switches. Encoding long interaction histories with standard attention has quadratic cost. This is impractical under strict online latency budgets. We introduce a Linear HSTU encoder. It replaces full attention with causal linear recurrence. Per-layer complexity drops from O(L^2) to O(L) with no loss in ranking quality. Traditional search ranking depends on hand-crafted features like text-matching scores, statistical signals, and behavioral features. We propose HFG-Attention to preserve them in the generative framework. It organizes heterogeneous features into semantic groups and fuses them through a dedicated attention block. This bridges sparse engineered signals with dense sequential representations. QGS is deployed in the ranking module of Quark Search, a major commercial search engine in China. Online A/B tests show statistically significant gains: +0.62% CTR, +0.38% Click-Search Ratio, and +3.55% PV Duration over the production deep learning baseline.
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。