先引用后发言：增强电商对话式 LLM 智能体中的上下文-响应落地
Cite Before You Speak: Enhancing Context-Response Grounding in E-commerce Conversational LLM-Agents
AI总结：
针对电商对话购物智能体的幻觉与来源不可验证问题，本文提出引用生成范式及 Multi-UX-Inference 系统，通过自动评估和在线 A/B 测试验证其可提升事实落地、客户信任与参与度。
摘要：
随着对话式大型语言模型（LLM）的发展，已有若干基于 LLM 的对话购物智能体（CSA）被开发出来，以帮助客户顺畅进行在线购物。构建具有吸引力且值得信赖的 CSA 的主要目标，是确保智能体关于产品事实型信息的回复准确且具有事实依据。然而，仍存在两个挑战。第一，LLM 会产生幻觉或无依据的声明。此类不准确内容可能传播错误信息并削弱客户信任。第二，如果 CSA 回复中没有提供知识来源归属，客户将难以验证 LLM 生成的信息。为同时解决这两个挑战，我们提出了一种易于产品化的解决方案，为客户提供“引用体验”。我们构建了自动评估指标，以全面评估 LLM 的落地和归属能力，结果表明引用生成范式将落地性能显著提升了 13.83%。为大规模部署该能力，我们引入了 Multi-UX-Inference 系统，该系统在 LLM 输出后附加来源引用，同时保留现有用户体验特性并支持可扩展推理。大规模在线 A/B 测试表明，依据事实落地的 CSA 回复可将客户参与度提升 3%–10%，具体取决于用户体验（UX）变体。
英文摘要：
With the advancement of conversational large language models (LLMs), several LLM-based Conversational Shopping Agents (CSA) have been developed to help customers smooth their online shopping. The primary objective in building an engaging and trustworthy CSA is to ensure the agent's responses about product factoids are accurate and factually grounded. However, two challenges remain. First, LLMs produce hallucinated or unsupported claims. Such inaccuracies risk spreading misinformation and diminishing customer trust. Second, without providing knowledge source attribution in CSA response, customers struggle to verify LLM-generated information. To address both challenges, we present an easily productionized solution that enables a ''citation experience'' to our customers. We build auto-evaluation metrics to holistically evaluate LLM's grounding and attribution capabilities, suggesting that citation generation paradigm substantially improves grounding performance by 13.83%. To deploy this capability at scale, we introduce Multi-UX-Inference system, which appends source citations to LLM outputs while preserving existing user experience features and supporting scalable inference. Large-scale online A/B tests show that grounded CSA responses improves customer engagement by 3% - 10%, depending on UX variations.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
