训练 LLM 与推荐系统混合模型：利用语义 ID 实现可引导推荐
原标题：Training an LLM-RecSys Hybrid for Steerable Recs with Semantic IDs
AI 摘要
Eugene Yan 训练了一个 LLM 与推荐系统的混合模型，通过语义 ID 将物品表示为语言模型可理解的 token，使模型既能基于行为数据推荐，又能通过自然语言对话进行引导和解释。实验使用 Amazon Reviews 2023 视频游戏数据，构建了 79k 用户序列，并采用 RQ-VAE 生成语义 ID。该混合模型虽在精度上不及专用推荐系统，但提供了可引导性和推理能力，统一了搜索、推荐和聊天体验。
正文节选
I got nerdsniped when I first heard about Semantic IDs. The idea is simple: Instead of using random hash IDs for videos or songs or products, we can use semantically meaningful tokens that an LLM can natively understand. I wondered, could we train an LLM-recommender hybrid on the rich behavioral data that makes today’s recommender systems so effective? To my surprise, we can! The result is a language model that can converse in both English and item IDs, not with retrieval or other tools, but as