提出可同时推理与推荐的统一模型，提升推荐效率与适应性。
R$^2$ec: Towards Large Recommender Models with Reasoning
- 双头架构支持推理链生成与高效物品预测
- 在三个数据集上超越传统及基于LLM的推荐模型
- 适合需要高适应性与低延迟推荐的场景
大型推荐模型通过编码或物品生成扩展了大语言模型的能力，近期大语言模型的推理突破也推动了推荐系统中推理能力的探索。本文提出R²ec，一种具备内在推理能力的统一型大推荐模型。R²ec采用双头架构，可在单一模型中同时支持推理链生成与高效物品预测，显著降低推理延迟。针对标注推理数据匮乏的问题，设计RecPO强化学习框架，通过新颖的融合奖励机制联合优化推理与推荐。在三个数据集上的大量实验表明，R²ec优于传统、基于LLM及增强推理的推荐基线；进一步分析验证其在常规基于LLM的推荐模型中具备竞争力的效率，且对多样化推荐场景具有强适应性。代码与模型检查点见https://github.com/YRYangang/RRec。
原文摘要 · Abstract (English)
Large recommender models have extended LLMs as powerful recommenders via encoding or item generation, and recent breakthroughs in LLM reasoning synchronously motivate the exploration of reasoning in recommendation. In this work, we propose R$^2$ec, a unified large recommender model with intrinsic reasoning capability. R$^2$ec introduces a dual-head architecture that supports both reasoning chain generation and efficient item prediction in a single model, significantly reducing inference latency. To overcome the lack of annotated reasoning data, we design RecPO, a reinforcement learning framework that optimizes reasoning and recommendation jointly with a novel fused reward mechanism. Extensive experiments on three datasets demonstrate that R$^2$ec outperforms traditional, LLM-based, and reasoning-augmented recommender baselines, while further analyses validate its competitive efficiency among conventional LLM-based recommender baselines and strong adaptability to diverse recommendation scenarios. Code and checkpoints available at https://github.com/YRYangang/RRec.
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。