OneRec-V2优化生成式推荐，提升效率与用户真实反馈对齐。
OneRec-V2 Technical Report
- 采用懒加载解码器架构，消除编码瓶颈，减少94%计算量。
- 通过时长感知奖励和自适应比例裁剪，提升用户行为对齐度。
- 在快手实测中显著提升停留时间，适合大规模生成推荐场景。
生成式AI的突破使推荐系统转向端到端生成范式。OneRec将推荐重构为自回归生成任务，实现高模型浮点运算利用率。尽管OneRec-V1在实际部署中表现优异，但其可扩展性和性能受两大挑战制约：（1）计算资源分配低效，97.66%的算力被序列编码占用而非生成；（2）强化学习依赖单一奖励模型。为此，我们提出OneRec-V2，包含：（1）懒加载解码器仅架构：消除编码器瓶颈，总计算量降低94%，训练资源节省90%，支持80亿参数规模扩展；（2）基于真实用户交互的偏好对齐：引入时长感知奖励塑造与自适应比例裁剪，更精准捕捉用户偏好。在快手平台的广泛A/B测试表明，OneRec-V2在保持多目标平衡的前提下，分别带来0.467%和0.741%的App停留时间提升。本工作推动了生成式推荐系统的可扩展性与真实反馈对齐，是端到端推荐系统发展的重要一步。
原文摘要 · Abstract (English)
Recent breakthroughs in generative AI have transformed recommender systems through end-to-end generation. OneRec reformulates recommendation as an autoregressive generation task, achieving high Model FLOPs Utilization. While OneRec-V1 has shown significant empirical success in real-world deployment, two critical challenges hinder its scalability and performance: (1) inefficient computational allocation where 97.66% of resources are consumed by sequence encoding rather than generation, and (2) limitations in reinforcement learning relying solely on reward models. To address these challenges, we propose OneRec-V2, featuring: (1) Lazy Decoder-Only Architecture: Eliminates encoder bottlenecks, reducing total computation by 94% and training resources by 90%, enabling successful scaling to 8B parameters. (2) Preference Alignment with Real-World User Interactions: Incorporates Duration-Aware Reward Shaping and Adaptive Ratio Clipping to better align with user preferences using real-world feedback. Extensive A/B tests on Kuaishou demonstrate OneRec-V2's effectiveness, improving App Stay Time by 0.467%/0.741% while balancing multi-objective recommendations. This work advances generative recommendation scalability and alignment with real-world feedback, representing a step forward in the development of end-to-end recommender systems.
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。