使用LLMs生成个人资料：了解消费者、商家和商品

To elevate the quality of personalization, DoorDash is evolving how we represent our core entities. For years, we have relied on embeddings — dense numerical vectors learned by deep neural network (DNN) models — to represent entities such as consumer, store, or item to power our search and recommendation system. While embeddings are compact and compute-friendly, they come with trade-offs: They are opaque to humans, difficult to debug, and hard to interpret directly. It is nearly impossible to explain to consumers and merchandisers why two vectors are similar — for example, “0.83 cosine distance.”

为了提升个性化的质量，DoorDash正在演变我们对核心实体的表示。多年来，我们依赖于嵌入——由深度神经网络（DNN）模型学习的密集数值向量——来表示消费者、商店或商品等实体，以推动我们的搜索和推荐系统。虽然嵌入是紧凑且计算友好的，但它们也有权衡：它们对人类是不可见的，难以调试，并且难以直接解释。几乎不可能向消费者和商品商解释为什么两个向量是相似的——例如，“0.83余弦距离。”

Large language models (LLMs) unlock a complementary representation — rich, narrative-style profiles written in natural language. These profiles preserve semantic nuance such as “prefers spicy Sichuan dishes, avoids dairy” for intuitive understanding, while remaining fully interpretable by humans. This not only allows us to build more intuitive, human-facing product features — for example, explaining why a dish is recommended — but also serves as powerful, structured input for next-generation LLM applications. It's how we're building a more explainable and semantically aware platform. More specifically, these LLM profiles enable:

大型语言模型 (LLMs) 解锁了一种互补的表示——用自然语言编写的丰富叙述风格的配置文件。这些配置文件保留了语义细微差别，例如“偏好辛辣的四川菜，避免乳制品”，以便直观理解，同时仍然完全可被人类解释。这不仅使我们能够构建更直观、面向人类的产品特性——例如，解释为什么推荐某道菜——而且还作为下一代 LLM 应用的强大结构化输入。这就是我们如何构建一个更具可解释性和语义意识的平台。更具体地说，这些 LLM 配置文件使得：

Transparent recommendations, such as “We suggested this dish because you liked X and it matches your pescatarian preference,”
透明的推荐，例如“我们建议这道菜，因为您喜欢 X，并且它符合您的海鲜偏好，”
Editable preferences that a customer support agent or the user can correct in plain English,
可编辑的偏好，客户支持代理或用户可以用简单英语进行更正，
Rapid feature prototyping so that product, marketing, or support teams can prompt LLMs agai...