Generative information retrieval (GenIR) formulates the retrieval process as a text-to-text generation task, leveraging the vast knowledge of large language models. However, existing works primarily optimize for relevance while often overlooking document trustworthiness. This is critical in high-stakes domains like healthcare and finance, where relying solely on semantic relevance risks retrieving unreliable information. To address this, we propose an Authority-aware Generative Retriever (AuthGR), the first framework that incorporates authority into GenIR. AuthGR consists of three key components: (i) Multimodal Authority Scoring, which employs a vision-language model to quantify authority from textual and visual cues; (ii) a Three-stage Training Pipeline to progressively instill authority awareness into the retriever; and (iii) a Hybrid Ensemble Pipeline for robust deployment. Offline evaluations demonstrate that AuthGR successfully enhances both authority and accuracy, with our 3B model matching a 14B baseline. Crucially, large-scale online A/B tests and human evaluations conducted on the commercial web search platform confirm significant improvements in real-world user engagement and reliability.

翻译：生成式信息检索（GenIR）将检索过程建模为文本到文本生成任务，借助大语言模型的丰富知识。然而，现有研究主要优化相关性，常忽视文档可信度。在医疗和金融等高价值领域，仅依赖语义相关性存在检索不可靠信息的风险。为此，我们提出权威感知生成式检索器（AuthGR）——首个将权威性纳入GenIR的框架。AuthGR包含三个关键组件：（i）多模态权威评分模块，利用视觉语言模型从文本与视觉线索中量化权威性；（ii）三阶段训练管线，逐步向检索器注入权威感知能力；（iii）混合集成管线，保障稳健部署。离线评估表明，AuthGR成功提升了权威性与准确性，其中3B参数模型性能对标14B基线模型。更重要的是，在商业网络搜索平台开展的大规模在线A/B测试与人工评估证实，该方法显著改善了真实用户交互体验与检索可靠性。