发表机构
Baidu; Beihang University; Renmin University of China（百度; 北京航空航天大学; 中国人民大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
本文针对现有电商搜索生成式检索方法的意图一致性问题，提出ICEGR框架，通过三个组件优化后，在离线指标及百度电商搜索A/B测试中均取得显著性能提升。
AI 中文摘要
生成式检索（GR）在电商搜索领域颇具应用前景，但现有方法在整个训练流程中难以保持查询意图一致性。其一，基于静态商品信息构建的语义ID（SID）限制了SID编码商品-意图关联的能力；其二，尽管监督微调（SFT）学习了全商品库的商品-SID映射，但低曝光商品仍缺乏真实查询意图监督，因为查询到SID的训练仅依赖在线日志，导致这些商品的检索性能较差；其三，面向业务的偏好优化可能会更倾向于热门或高价值商品，而非最匹配查询意图的商品，削弱了查询-商品相关性。为解决这些问题，本文提出ICEGR，这是一种在GR训练流程中始终整合查询意图的面向电商搜索的意图一致型端到端生成式检索框架。ICEGR包含三个组件：（1）意图感知型SID构建，将查询意图信号整合到SID构建过程中，使SID能够捕获静态商品信息之外的搜索意图；（2）合成查询增强型统一SFT，在查询到SID的目标下统一多个SFT任务，并利用合成查询扩充在线日志提供的稀疏监督，为低曝光商品提供互补的查询意图监督；（3）相关性校准型偏好优化，将查询-商品相关性与业务信号整合到边际自适应偏好目标中，在保留查询意图的同时实现业务偏好学习。离线实验结果显示，ICEGR相比基线方法将Recall@20提升了21.7%，NDCG@20提升了26.6%；作为端到端生成式检索路径部署于百度电商搜索后，A/B测试中ICEGR实现了点击率（CTR）相对提升3.52%，订单量相对提升15.96%，商品交易总额（GMV）相对提升7.53%。
英文摘要
Generative Retrieval (GR) is promising for e-commerce search, yet existing methods struggle to maintain query-intent consistency throughout the training pipeline. First, semantic ID (SID) construction based on static product information limits the ability of SIDs to encode product-intent associations. Second, although supervised fine-tuning (SFT) learns product-SID mappings across the catalog, low-exposure products still lack real query-intent supervision because query-to-SID training relies solely on online logs, resulting in poor retrieval performance for these products. Third, business-oriented preference optimization may favor popular or high-value products over those that best match the query intent, weakening query-product relevance. To address these issues, we propose ICEGR, an Intent-Coherent End-to-End Generative Retrieval Framework for E-commerce Search that integrates query intent consistently throughout the GR training pipeline. ICEGR comprises three components: (1) Intent-Aware SID Construction incorporates query-intent signals into SID construction, enabling SIDs to capture search intent beyond static product information; (2) Synthetic Query-Enhanced Unified SFT unifies multiple SFT tasks under the query-to-SID objective and augments sparse supervision from online logs with synthetic queries, providing complementary query-intent supervision for low-exposure products; and (3) Relevance-Calibrated Preference Optimization integrates query-product relevance and business signals into a margin-adaptive preference objective, preserving query intent while enabling business preference learning. Offline results show that ICEGR improves Recall@20 by 21.7% and NDCG@20 by 26.6% over the baseline. Deployed as an end-to-end generative retrieval pathway in Baidu E-commerce Search, ICEGR achieves relative improvements of 3.52% in CTR, 15.96% in order volume, and 7.53% in GMV in an A/B test.
Comments12 pages, 5 figures
