让推荐模型像人一样思考，提升个性化推荐的透明度与效果
OneRec-Think: In-Text Reasoning for Generative Recommendation
- 引入显式推理机制，让推荐过程可解释
- 在多个数据集上达到领先性能，工业部署提升停留时长0.159%
- 适合追求可解释推荐和落地应用的研究者与工程师
大型语言模型（LLM）的强大生成能力正推动推荐系统范式变革。然而，现有生成式模型（如OneRec）作为隐式预测器，缺乏显式可控的推理能力——这正是LLM的核心优势。为此，我们提出OneRec-Think，一个融合对话、推理与个性化推荐的统一框架。该框架包含：(1) 项目对齐（Itemic Alignment）：跨模态项目-文本对齐，实现语义锚定；(2) 推理激活（Reasoning Activation）：通过推理支架激活推荐场景下的LLM推理能力；(3) 推理增强：设计考虑用户偏好多有效性特征的推荐专属奖励函数。在公开基准上的实验表明其达到当前最优性能。此外，所提出的“Think-Ahead”架构已在快手实现有效落地，使APP平均停留时长提升0.159%，验证了显式推理能力的实际价值。
原文摘要 · Abstract (English)
The powerful generative capacity of Large Language Models (LLMs) has instigated a paradigm shift in recommendation. However, existing generative models (e.g., OneRec) operate as implicit predictors, critically lacking the capacity for explicit and controllable reasoning-a key advantage of LLMs. To bridge this gap, we propose OneRec-Think, a unified framework that seamlessly integrates dialogue, reasoning, and personalized recommendation. OneRec-Think incorporates: (1) Itemic Alignment: cross-modal Item-Textual Alignment for semantic grounding; (2) Reasoning Activation: Reasoning Scaffolding to activate LLM reasoning within the recommendation context; and (3) Reasoning Enhancement, where we design a recommendation-specific reward function that accounts for the multi-validity nature of user preferences. Experiments across public benchmarks show state-of-the-art performance. Moreover, our proposed "Think-Ahead" architecture enables effective industrial deployment on Kuaishou, achieving a 0.159\% gain in APP Stay Time and validating the practical efficacy of the model's explicit reasoning capability.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。